Facebook’s parent company has begun notifying staff worldwide that they are out of a job, with engineers and product teams bearing the brunt of a 10 per cent cull designed to bankroll a $145bn artificial intelligence spending spree.

Meta Platforms started handing out redundancy notices on Wednesday morning, kicking off one of the most aggressive restructurings in Silicon Valley this year. As many as 8,000 roles, roughly a tenth of the company’s global headcount, are expected to disappear as Mark Zuckerberg shifts the business onto a leaner, AI-first footing.

The cuts are heavily concentrated in the company’s engineering and product divisions, according to a Bloomberg report, with around 350 jobs in Dublin, Meta’s European headquarters, set to go. The Irish capital has long been a critical hub for the owner of Facebook, WhatsApp and Instagram, hosting thousands of staff serving customers across the EMEA region.

Even before the redundancy letters landed, the wheels of internal change were already in motion. On Monday, some 7,000 employees were told they had been redeployed to newly formed teams charged with developing AI products, agents and assistants that will be threaded through Meta’s family of apps.

“We’re now at the stage where many orgs can operate with a flatter structure with smaller teams of pods/cohorts that can move faster and with more ownership,” Janelle Gale, Meta’s chief people officer, wrote in an internal memo seen by staff this week.

A $145bn bet on ‘personal superintelligence’

The job losses come as Meta pours unprecedented sums into the data centres, chips and engineering talent it believes will define the next decade of computing. At its most recent quarterly results, the company told investors it would spend up to $145bn on capital expenditure this year, more than double the $72bn it shelled out in 2025.

Where rivals such as Google, Microsoft and Amazon are funnelling much of that AI capability into cloud services they can sell to corporate customers, Mr Zuckerberg is taking a different path. The Meta co-founder is pursuing what he calls “personal superintelligence” — a hyper-personalised AI assistant designed to live inside Facebook, Instagram, WhatsApp and the company’s growing range of smart glasses and headsets.

Meta’s Muse Spark model, released in April, is the first significant product to emerge from its Superintelligence Labs unit, which was set up last June and stocked with high-profile hires poached from OpenAI, Anthropic and Google DeepMind.

That spending has unnerved investors and weighed on the share price. Meta’s stock is down 8.4 per cent so far this year, even as the wider Nasdaq has put on 12.5 per cent, a divergence that, as Business Matters reported after the first-quarter results, reflects mounting unease over the lack of a direct revenue line attached to Meta’s AI bill. When pressed on the return on investment of the spending, Mr Zuckerberg told analysts on the Q1 earnings call that it was “a very technical question”, a line that did little to soothe nerves on Wall Street.

A wider AI-driven shake-out in tech

Meta is far from alone in trying to wring efficiencies out of its workforce while throwing money at AI. Intuit, the American owner of QuickBooks and TurboTax, is preparing to lay off around 17 per cent of its workforce, or roughly 3,000 staff. Amazon, Microsoft, Cloudflare and Jack Dorsey’s payments group Block have all announced major redundancy rounds this year, with Amazon’s own 16,000-job cull framed by chief executive Andy Jassy as a way to “remove bureaucracy”.

According to Layoffs.fyi, which tracks redundancies in the tech sector, more than 140 companies have laid off in excess of 111,000 employees so far this year, already closing in on the 124,636 cuts recorded across the whole of 2025.

For UK small and medium-sized businesses, the message from the world’s most valuable technology companies is unmistakable. Capital that once funded sprawling product teams is now being redirected into infrastructure, models and a much smaller pool of senior engineers. As consultancy giants such as McKinsey trim their own ranks on the same logic, British SME owners weighing their own AI strategies face an uncomfortable question: are they investing fast enough to keep up, or being lured into a costly arms race they cannot win?

Meta and Intuit were contacted for comment.

Read more:
Meta to axe 8,000 jobs as Zuckerberg doubles down on AI race

You may have heard the joke: an older fish says to a younger fish, ‘The water’s nice today, huh?’ and the younger fish replies, ‘What the hell is water?’ It works because the things that shape our experience most are often the easiest to overlook.

For many businesses, workforce technology is like that. When it works properly, nobody notices it. When it doesn’t, it can quickly become the centre of the working day. And if you don’t have a dedicated IT team to step in and fix issues quickly, the impact is magnified.

This is felt especially sharply with employee laptops, because so much of modern work runs through this single device: email, documents, spreadsheets, browser tools, calls, messaging and client communication. When a laptop is not up to the job, it reshapes how work feels, how smoothly people move through the day and how much energy gets wasted on things that should be effortless. Crucially, this often doesn’t show up as one dramatic failure. It shows up as constant, low-level friction that people gradually learn to work around. That is what makes it so easy to miss. Employees adapt, lower expectations, build bad habits to cope with the device and push through, so the drag on time and energy becomes ‘just how it is’.

In practice, that can mean slowdowns when switching between email, documents, spreadsheets, browser tabs and calls, video meetings that glitch, freeze or feel unreliable under pressure, battery anxiety when working away from a desk, repeatedly waiting for the laptop to catch up, restart or reconnect, cramped side-by-side working on smaller screens, and too much reliance on dongles, adapters and setup workarounds.

The cost in terms of behavioural impact includes employees switching cameras off just to keep calls running smoothly, which hampers communication and damages the client experience, keeping fewer windows open than they need, which slows tasks down, delaying restarts and important software updates, increasing exposure to vulnerabilities, and using their personal devices as a backup, sometimes handling sensitive business or customer information.

For business leaders, there is another layer of concern: buying the wrong thing and being stuck with it for years. That might mean devices already feeling stretched after 12 to 24 months, overspending on tech people do not fully use, or risking client trust through weak privacy and security. The biggest risk is that these ways of working start to feel normal. Once that happens, friction stops looking fixable and starts getting absorbed into everyday life.

Because people often stop flagging these issues and simply work around them, it’s easy for leaders to underestimate the scale of the problem. But this is affecting millions of SMBs in the UK and many millions more around the world. HP’s 2026 SMB workflow research found that nearly 60% of SMB IT leaders say troubleshooting consumes more of their time than innovation, nearly half of SMB workers say obsolete tools make everyday tasks unnecessarily frustrating, and more than 60% of business leaders link those inefficiencies to increased burnout and employee turnover.

If hidden friction is the problem, then simply adding more technology is not the answer. Rather, it is about how to choose the right devices that will remove the most important points of friction from the working day.

The HP EliteBook 8 G1a, advanced by the AMD Ryzen AI 7 Pro is a useful example of a lower-friction device because it is built around the problems businesses actually experience. Work feels faster and less stop-start, because the laptop has the headroom for how people actually work now, moving between documents, spreadsheets, browser tabs, messaging and HD calls without quickly feeling maxed out. That is where the AMD Ryzen AI 7 Pro platform, 64GB RAM and 1TB storage make a real difference.

Long, multitasking sessions feel more comfortable, because the 16-inch, 16:10 display gives people more room to compare documents, work across spreadsheets and take notes during meetings without constant resizing and juggling. Hybrid work becomes less awkward, because built-in HDMI, USB-A and multiple USB-C and Thunderbolt 4 ports make it easier to move between meeting rooms, home offices and shared workspaces without relying on a bag full of dongles and adapters.

Security and privacy feel more built in and less disruptive, which matters especially for SMBs without a dedicated IT team. HP Wolf Security helps isolate common threats such as phishing links, malware and ransomware in the background, while Sure View narrows the viewing angle of the screen so sensitive information is harder for people nearby to see in shared or public spaces. Meetings feel more professional without extra effort, because the 5MP camera and built-in AI-powered meeting features help people look clear, stay centred in frame and sound better on calls.

As a next generation AI PC, it is a more future-ready choice, because AI will increasingly be part of the tools businesses already use. With a dedicated Neural Processing Unit (NPU) and enough memory to support more local AI-enabled workloads over time, it is designed to stay fast and efficient for longer rather than feeling like the wrong decision a year from now.

For business leaders, the key question is: What will reduce friction for our team for long enough to justify the investment? Some useful ways to think about this, and questions to ask your team directly, include identifying where current laptops are quietly slowing people down, looking for repeated low-level problems rather than dramatic failures such as lag, poor meetings, awkward setup, battery stress and too many workarounds. It also means understanding what the busiest day actually looks like and buying for the reality of multitasking, video calls, side-by-side working and hybrid movement.

Leaders should consider whether they are buying for short-term savings or long-term value, since a cheaper device that feels stretched after a year can become worse value than a better-specced one that stays comfortable for longer. They should also ask whether security feels built in or bolted on, because the safest setup is usually the one that asks the least extra effort from already busy people.

It is also worth thinking about whether a device will stay useful as AI-enabled tools become more normal. The practical issue is not whether AI matters this minute, but whether the laptop will keep pace as those features become part of everyday software. Finally, consider whether the device fits how people actually work, as the right choice is about balance: performance headroom, screen space, connectivity, collaboration and peace of mind.

Future-ready technology should not demand more attention from a business. It should support the business without demanding more effort to use it, by reducing everyday friction, protecting sensitive work and staying useful for long enough to offer real value. For more information, please visit HP’s site.

Read more:
Are You Building a Future-Ready Business? Choose Tech That Is Less Visible, Not More Complicated

Britain’s small and medium-sized businesses have been put on notice. From 19 June 2026, exactly one month from today, every organisation that handles personal data will, by law, be required to operate a formal complaints process. Those that fail to prepare risk regulatory action, reputational damage and the slow drip of customer trust eroding away.

The new obligations flow from section 103 of the Data (Use and Access) Act 2025, the most significant reshaping of the UK’s data protection landscape since the post-Brexit settlement. And in a clear signal that the Information Commissioner’s Office is anxious to avoid a repeat of the GDPR scramble of 2018, deputy commissioner Emily Keaney has used the four-week countdown to issue a direct appeal to the smaller end of the market.

“There is still plenty of time to act, and the ICO is here to support you,” Ms Keaney said. “We know that smaller organisations are less likely to have formal complaints processes in place, and that is exactly why we have designed this guidance with you in mind.”

What the new law actually requires

For SME owners and finance directors who have not yet digested the detail, the statutory obligations are mercifully short. Under the new regime, every organisation must give individuals a clear and accessible route to raise a data protection complaint, whether by email, online form, telephone or post. Receipt of a complaint must be acknowledged within 30 days. Businesses must then, “without undue delay”, take appropriate steps to investigate, keep the complainant informed of progress, and communicate the outcome.

Crucially, there are no carve-outs. The rules apply to the corner shop with a customer mailing list just as much as to the FTSE 250 financial services firm. Privacy notices will also need updating to make clear that customers have a right to complain directly to the organisation before escalating to the regulator.

Why this matters more than it might look

On paper, the changes appear modest, a tweak to administrative housekeeping rather than the seismic shock that GDPR delivered seven years ago. But seasoned compliance professionals warn that complacency would be a mistake.

For the first time, individuals will have a statutory right to complain directly to the organisation handling their data, and to expect a structured response within a defined timeframe. That changes the calculus on everything from subject access requests to the handling of data breaches. The ICO has indicated that sectors generating the highest volume of complaints, healthcare, financial services, technology and retail, should expect particular scrutiny.

There is also a commercial logic at work. Resolving a grievance quickly and fairly tends to prevent it from metastasising into something more serious, whether a formal regulatory referral or a customer departure. As any SME operator who has watched a one-star Trustpilot review go viral can attest, the cost of getting the response wrong can dwarf the cost of getting the process right. The wider context is one of rising data risk, with the ICO already pressing the technology sector to embed privacy by design into AI products, a sign of how high the regulatory bar is climbing.

The ICO’s olive branch

The regulator’s tone this time is markedly different from the rather schoolmasterly approach that characterised the early GDPR rollout. The guidance, published in February following a public consultation that drew more than 85 responses, is studded with practical examples and worked-through scenarios pitched squarely at smaller firms without dedicated compliance teams.

“A data protection complaint can come from any customer at any time,” Ms Keaney noted. “Having a clear process means you can respond quickly, resolve issues fairly and protect the trust your customers place in you. We are not here to catch businesses out, we are here to help you get ready.”

That conciliatory framing should not, however, be mistaken for indefinite patience. Once the 19 June commencement date passes, the ICO will have the power to take enforcement action against organisations that fail to operate a compliant process, and the line between supportive regulator and active enforcer can move quickly.

A four-week action list

For business owners still unsure where to begin, the practical steps are reasonably straightforward. Decide who inside the business will own the complaints process and ensure they have the authority to investigate and respond. Build a simple, visible route for customers to raise complaints — usually a dedicated email address or web form, signposted in the privacy notice. Document the workflow, including how the 30-day acknowledgement deadline will be met. Train any customer-facing staff on what to do if a complaint lands in their inbox.

Owners who already operate under data protection frameworks will recognise much of this from existing good practice. For a refresher on the broader compliance landscape, our complete guide to GDPR compliance in the UK sets out the foundations, while our explainer on the difference between data controllers and processors is worth bookmarking for any business that shares customer data with third parties.

The bottom line

For Britain’s 5.5 million SMEs, the message from regulators is clear: 19 June is not a target, it is a deadline. The four weeks ahead are not an invitation to delay, but a window to prepare. Done well, the new complaints process is a modest piece of administrative plumbing that can quietly strengthen customer relationships. Done badly, or not at all, it is a regulatory exposure that few small businesses can afford to carry.

The ICO has, unusually, all but rolled out a welcome mat. The smart move for SME owners is to walk through the door before someone else knocks.

Read more:
ICO Warns SMEs: one month to comply with new Data Complaints Law

An Isle of Man trading-education platform has won a two-year trade mark battle against TikTok’s UK arm, in a ruling small business advisers say sets a powerful precedent for founders facing legal pressure from global tech giants.

In a decision likely to be studied across the SME community, a small financial-trading education business has seen off TikTok Information Technologies UK Limited, the British arm of ByteDance, one of the world’s most valuable technology companies, in a two-year trade mark dispute before the UK Intellectual Property Office (UKIPO).

TickTickTrader Ltd, an Isle of Man-based platform that trains aspiring futures traders, applied in April 2023 to register the mark ‘TickTickTrader’. The name is drawn from trading floor terminology: a ‘tick’ is the smallest permissible price movement on a futures exchange. TikTok’s legal team, drawn from one of the world’s largest law firms, opposed the application outright, arguing that the name was confusingly similar to its own globally recognised brand and risked diluting its reputation. An accompanying cease-and-desist letter gave the start-up 14 days to withdraw the application, abandon the name and sign undertakings.

For a company of around 50 employees, the demand was existential. It refused.

On 19 February 2026, Hearing Officer Mrs E Fisher dismissed both grounds of TikTok’s opposition in full and ordered TikTok to pay TickTickTrader’s costs. The appeal window has now closed and the decision is final.

‘Tick Tick’ is not ‘Tik Tok’

The Hearing Officer found the two marks to be visually and aurally similar only to a medium degree, and, crucially, conceptually dissimilar. Where TikTok evokes the sound of a clock, TickTickTrader was held to conjure the image of a trader methodically ticking off positions, gain by incremental gain. The word ‘trader’, the officer ruled, was neither irrelevant nor purely descriptive: it formed an integral part of the overall impression of the mark.

“I find there is no likelihood of direct or indirect confusion,” the decision states. The officer also rejected TikTok’s argument that consumers would assume TickTickTrader was a brand extension in the same family as TikTok Shop, TikTok Pay or TikTok Live, describing that logic as unconvincing. “I find that there is no link between the marks,” she wrote.

She added that buyers of education and training services, typically high-attention, considered purchases, were highly unlikely to confuse two marks with such clear visual, conceptual and commercial differences. TikTok’s reputation, however considerable, did not entitle it to monopolise the market.

Big tech, small business – and the cost of standing your ground

TickTickTrader fought the opposition with Trade Mark Wizards, the London-based intellectual property firm backed by Lord Sugar, who is also a director of the company. For Trade Mark Wizards, the case is emblematic of a wider pattern in which large corporations rely on the disproportionate commercial pressure of legal proceedings to push smaller rivals into surrender, regardless of the underlying merits.

It is a pattern this magazine has documented before, from Rolex demanding that a Devon children’s clock start-up change its name to the steady stream of cease-and-desist letters dropped on UK founders by global brands. As Business Matters has previously argued in its guidance for founders accused of trade mark infringement, not every claim is legally sound, and capitulating without a proper assessment can prove far more costly than fighting back.

Lord Sugar, director at Trade Mark Wizards, said the TickTickTrader case carried a familiar shape.

“I’ve been in business long enough to recognise this pattern straight away,” he said. “Big companies think they can throw their weight around and that smaller businesses will just roll over because they can’t afford the fight. That’s not how it’s supposed to work. What mattered here is that the claim didn’t stack up — and when it was properly tested, it failed. You don’t get to own every name that sounds vaguely similar to your own just because you’re a big brand. If you’re right, you’re right. If you’re not, you lose. Simple as that.”

Oliver Oguz, managing director of Trade Mark Wizards, was equally direct. “The playbook used to be simple,” he said. “If you’re a big company and a small business gets in your way, you throw lawyers at it and wait for them to blink. That playbook is finished. This decision is proof that the rules apply to everyone, regardless of how many zeros are in your legal budget. TikTok had every resource in the world at its disposal. They still lost because the facts didn’t support them.”

A board member of TickTickTrader described the moment the ruling came through.

“We were effectively being asked to give up our brand entirely. For a small business, that’s not just a legal issue, it’s your identity, your work, everything you’ve built. It would have been easy to walk away, but we knew the name meant something and we believed we were right to keep it. Having the right support around us made all the difference.”

What founders can take from the ruling

For start-ups and SMEs watching from the sidelines, the case offers three practical lessons. First, the UKIPO’s standard opposition process is structured, evidence-led, and decided on the law — not on the relative size of the parties. Second, a defence grounded in the genuine meaning and commercial context of a brand name can defeat a much larger opponent. Third, as our own legal contributors have long argued in pieces on brand protection and the power of IP, independent legal advice from specialists, rather than reflexive capitulation, is often the decisive factor in determining whether a name survives.

The UKIPO has confirmed that TickTickTrader may now proceed to full trade mark registration, which the company has also secured in several key global territories. TikTok has been ordered to pay £1,700 in costs. The figure is modest in headline terms, but reflects the tribunal’s clear view on the merits, and, for the wider SME community, the symbolism is anything but small.

Read more:
How a 50-person start-up beat TikTok at the IPO – with Lord Sugar in its corner

Britain’s artificial intelligence sector has produced its first heavyweight league table of 2026, with Barclays placing Oxford-founded chip designer Fractile and Google DeepMind spinout Isomorphic Labs at the centre of its new AI 100 ranking, a list that crystallises just how quickly the UK’s AI economy is maturing.

The bank’s Eagle Labs division, the high-street lender’s start-up incubator network, unveiled the inaugural ranking this week to spotlight the country’s fastest-growing AI businesses. Its publication coincides with what is shaping up to be a record year for the sector, with UK AI companies hoovering up £8.3bn of investment in 2025 alone and cementing London’s status as Europe’s most prolific AI capital.

For Britain’s policymakers, under pressure to deliver on the Prime Minister’s pledge to “mainline AI into the veins” of the economy, the league table arrives at a politically charged moment. For investors, it offers a useful shortlist of the companies global capital is now chasing hardest.

Oxford chip pioneer joins the unicorn club

Few names on the ranking have captured boardroom attention quite like Fractile. The Oxford-founded business, set up in 2022 by former university researcher Walter Goodwin, this week banked a $220m (£165m) Series B led by Peter Thiel’s Founders Fund, with Accel and Factorial Funds joining the cheque.

The round vaults Fractile into the so-called unicorn bracket and underlines a belief among Silicon Valley’s most influential investors that the next great AI bottleneck will not be cleverer algorithms, but the eye-watering cost of running them. Mr Goodwin’s firm is racing to build inference chips that promise to slash the price of deploying AI models at commercial scale, a problem that has come to dominate boardroom conversations from Wall Street to Whitehall.

Industry watchers say the deal is one of the clearest signals yet that British deep-tech, long accused of losing its champions to American buyers, can hold its own on global capital markets. It also lands at a moment when Westminster is leaning heavily on the semiconductor sector to underpin its growth narrative, having earlier expanded backing for chip start-ups through the ChipStart programme.

Isomorphic eyes a pharma revolution

If Fractile represents the picks-and-shovels end of the AI gold rush, Isomorphic Labs sits at the other extreme. The London-based drug-discovery business, spun out of Google DeepMind in 2021 under the stewardship of Sir Demis Hassabis, recently sealed a $2.1bn (£1.57bn) funding round, one of the largest AI raises seen in Europe to date.

The company is using machine learning to accelerate the early-stage development of new medicines, an area where pharmaceutical giants have spent years grappling with stubbornly long timelines and ballooning research budgets. Big Pharma is already paying attention: AstraZeneca and Eli Lilly have inked partnerships, and a maiden in-house drug candidate is expected to enter clinical trials before the end of the year.

For an industry where the average new medicine takes more than a decade and over $2bn to bring to market, the prospect of AI compressing that timeline is no longer theoretical. It is precisely the sort of productivity dividend that researchers at HSBC say could deliver a £105bn revenue uplift to Britain’s mid-sized firms by 2030 if AI adoption keeps pace.

A boom under scrutiny

Yet for all the bullish numbers, Britain’s AI investment surge is not without its sceptics. A recent investigation by the Guardian questioned whether several headline-grabbing pledges promoted by ministers — including data-centre commitments linked to Nvidia-backed groups Nscale and CoreWeave, had been overstated.

The newspaper reported that some projects billed as brand-new infrastructure were in reality expansions of existing facilities. The Department for Science, Innovation and Technology (DSIT) rejected the bulk of the claims but conceded it was “not playing an active role in auditing these commitments”.

The episode is symptomatic of a broader credibility test now facing governments worldwide as they trumpet AI as the engine of future growth. The UK has so far announced a £500m Sovereign AI Unit and additional billions of pounds in compute and infrastructure spending, but ministers are increasingly being asked to demonstrate that the eye-catching figures translate into real jobs, factories and tax receipts.

A maturing market

Even so, the trajectory looks unmistakable. With more than £8bn raised across the sector last year, five fresh unicorns minted and at least 67 exits worth a combined £4bn, the British AI ecosystem is no longer trading on potential alone. Smaller players are also benefiting: Eagle Labs’ broader incubator network, which has supported thousands of regional start-ups through schemes such as its £12m regional grant programme, is increasingly being used as a pipeline-builder for the next cohort of AI 100 candidates.

For Barclays itself, the ranking is a useful piece of brand-building among the founders it hopes to bank for years to come. For Britain, it is something rather more consequential, an early glimpse of the companies that may, within a decade, sit alongside the country’s established corporate giants.

As one venture capitalist put it this week: “Five years ago, you’d struggle to name three UK AI businesses worth backing. Today you can’t fit them on a single page.” On the strength of Barclays’ latest list, that problem is unlikely to disappear any time soon.

Read more:
Barclays crowns Fractile and Isomorphic Labs in inaugural AI 100 as Britain’s tech race intensifies

For all the hand-wringing over privacy, Britain’s high streets, gyms and offices are about to be flooded with cameras hiding in plain sight.

The latest generation of so-called smart glasses, most notably Meta’s Ray-Ban range, has become one of the fastest-selling consumer electronics products in history, and the world’s largest technology companies are queuing up to follow suit.

The commercial momentum is undeniable. Meta has now shipped more than seven million pairs of its Ray-Ban smart glasses, made in partnership with Franco-Italian eyewear giant EssilorLuxottica, and the device accounts for more than 80 per cent of the global AI eyewear market, according to Counterpoint Research. Mark Zuckerberg, Meta’s chief executive, told investors earlier this year that the glasses were “some of the fastest-growing consumer electronics in history”, a rare bright spot for a company that has spent tens of billions of dollars chasing the metaverse with limited return.

But the same product line is now sitting at the centre of a rapidly widening privacy row that could shape regulation, workplace policy and consumer trust for years to come, and which British SMEs, from beauty salons to cafés, are already being forced to think about.

A camera in every frame

The appeal of the device, on paper, is straightforward. The Ray-Ban model carries an almost invisible camera in the frame, small open-ear speakers in the arms, and a discreet indicator light. Wearers can take a photo, capture video, place a phone call or summon Meta’s AI assistant with a tap on the temple. For early adopters such as Mark Smith, a partner at advisory firm ISG, the attraction is mundane rather than futuristic. He wears his every day, he says, because they let him take a call or listen to a podcast while washing up without blocking out the room, and spare him from pulling out a phone to capture a moment while travelling.

The problem, as Smith himself concedes, is that nobody around the wearer can tell. The recording light is dim in daylight and easily missed. To the casual observer, the glasses look like any other pair of Wayfarers.

That ambiguity is now generating an uncomfortable run of headlines. Women have reported being approached on beaches, in shops and on the street by men wearing the glasses, who film their reactions to scripted pick-up lines or intrusive questions and then upload the clips for clicks. Victims often only discover the footage exists once it has gone viral — and any subsequent abuse with it. As photography in public places is broadly lawful in the UK, legal recourse is limited. One woman who asked for her secretly recorded video to be removed told the BBC she was informed by the poster that takedown was “a paid service”.

Lawsuits, content moderators and a Kenyan flashpoint

The reputational pressure on Meta has been compounded by the working conditions of those who train the AI behind the product. Content moderators in Kenya, tasked with reviewing footage captured through the glasses to build training data, alleged they had been required to watch graphic material including sexual activity and people using the lavatory. Two lawsuits followed from owners of the glasses themselves: one group claiming they had no idea such videos had ever been captured, another that they had not realised the footage was being shared back to Meta for human review.

The company has pointed to its terms of service, arguing that the possibility of human review in certain circumstances had been disclosed. A Meta spokesman, Tracy Clayton, told the BBC: “We have teams dedicated to limiting and combating misuse, but as with any technology, the onus is ultimately on individual people to not actively exploit it.”

That defence is unlikely to satisfy the regulators now circling the category. Meta is reportedly preparing to add facial recognition to a forthcoming version of the glasses, according to The New York Times, a feature that would allow wearers not just to record passers-by, but to identify them in real time.

The rest of Silicon Valley piles in

For all the controversy, the rest of Big Tech sees a market it cannot afford to miss. Apple is widely reported to be developing its own smart glasses, with Bloomberg suggesting a launch as soon as next year. Snap has confirmed a new, lighter pair of its Specs for 2026. Google, more than a decade on from the spectacular failure of Google Glass, pulled within two years of launch amid a furious privacy backlash, is preparing another attempt under its Android XR platform.

Analysts at Citigroup and researchers at UC Berkeley reckon as many as 100 million people could be wearing AI-enabled glasses within a few years. For investors, that points to a genuinely new product category, the first since the smartwatch. For regulators, public bodies and small businesses, it raises a far thornier question: how do you enforce existing rules against recording in courtrooms, hospitals, changing rooms, museums, cinemas and bathrooms when a meaningful slice of the population is wearing a camera on their face?

David Kessler, who leads the US privacy practice at international law firm Norton Rose Fulbright, says corporate clients are already wrestling with it. “There are some pretty dark places we could go here,” he said. “I’m not anti-technology in any sense, but as a societal matter… will I need to think [of being recorded] anytime I go out in public?”

What it means for British SMEs

For owner-managers in the UK, this is no longer a Silicon Valley curiosity. Anecdotes are mounting of customers and staff being caught off guard: the online influencer Aniessa Navarro recounted feeling “sick” when she realised mid-treatment that her beauty technician was wearing Meta’s glasses. The technician insisted they were neither charged nor recording, and were needed for prescription lenses — but the reputational risk for the salon is obvious.

Smaller businesses in hospitality, retail, healthcare, fitness and personal services should expect to revisit their acceptable-use policies, customer-facing signage and staff training. Under UK GDPR, covert recording of identifiable individuals on a business’s premises is likely to fall on the operator as well as the wearer once that footage is processed for any purpose beyond purely personal use. Insurers and trade bodies are likely to start asking questions.

Meta markets the product under the tagline “Designed for privacy, controlled by you”, and tells wearers not to record people who object and to switch the glasses off entirely in sensitive spaces. Those suggestions, by the company’s own admission, are honoured more in the breach than the observance. A growing genre of “prank” content sees young men in Ray-Bans persuading retail workers to smell candles laced with foul odours, getting members of the public to sign fake petitions, or filming themselves snatching food at drive-throughs.

A Google Glass moment, or a tipping point?

Andrew Bosworth, Meta’s chief technology officer, was asked on Instagram about “the stigma around people wearing smart glasses every day”. His answer leant heavily on the sales figures, arguing that the sheer volume shifted “suggest these are widely accepted”.

Not everyone is convinced. David Harris, a former Meta AI researcher now teaching at UC Berkeley and advising policymakers in the US and EU, believes the category is heading for the same wall that flattened Google Glass. “Technology like this is fundamentally an invasion of privacy and it’s really going to face more and more backlash,” he said.

The signs are already there. In December, a New York man posted a clip lamenting that a woman he had been filming on the subway had broken his Meta glasses. The internet did not commiserate. It crowned her a folk hero.

For Meta, for Apple, for Snap and for Google, the commercial prize from owning the face is enormous. But for an industry that has spent the past decade trying to rebuild public trust, betting the next platform on a device most bystanders cannot tell is a camera may yet prove the most expensive miscalculation of all.

Read more:
Smart glasses are ‘an invasion of privacy’, yet Meta is shifting them by the million

Amazon has quietly opened a new front in the battle for ultra-fast delivery, becoming the first retailer in Britain to drop parcels by drone after a limited launch in Darlington, County Durham.

The service, operated under the company’s long-gestating Prime Air programme, will see packages weighing less than 5lb (2.2kg) flown out from an Amazon fulfilment centre to homes within a 7.5-mile (12km) radius. Initial payloads are unglamorous but practical: beauty products, batteries, charging cables and the kind of small household items shoppers tend to discover they need only when it is already too late to drive to the shops.

For Amazon, which first promised drone deliveries more than a decade ago and has since watched the technology stutter through regulatory and engineering setbacks, the Darlington launch is both a proof point and a test bed. For Britain’s retail sector, including the small and medium-sized businesses that increasingly rely on Amazon’s logistics network, it is a sharper reminder still that the goalposts on customer expectation are moving once again.

The trial’s earliest beneficiary was Rob Shield, a Darlington farmer who let Amazon use an Airbnb on his land for its first test runs. The novelty, he admits, soon took over.

“Initially it was a novelty, so we were ordering everything under the sun,” he says. “Pens, paper, chocolates, anything to make it keep coming.”

Parcels arrive in shoebox-sized packages, released from a height of around 12ft onto the front garden. The spectacle, Mr Shield concedes, drew its own audience: “We’d have people come just to see it.”

What began as a curiosity has become, in his telling, a quiet utility. “You start realising, ‘I actually need something today’, like tape measures and stuff you’re always losing. We just order it and it comes.”

In the UK, Amazon’s drones currently promise delivery within two hours. The American benchmark is rather more pointed: David Carbon, vice president of Amazon Prime Air, says the average delivery time in the US is now 36 minutes.

“The certainty is people have never told us they want their stuff slower,” he says. “If you’ve got kids and you want fever medication, you want it. You don’t want to drive to the store.”

Amazon will cap operations at ten flights an hour and up to one hundred deliveries a day on weekdays, a deliberately modest cadence designed to satisfy regulators rather than sceptical shareholders.

The aircraft in question is the MK30, Amazon’s latest model, fitted with sensors intended to avoid trampolines, washing lines, pedestrians and other aircraft. GPS guides the drone to each drop-off, where it releases its load. “This is effectively an autonomous drone that can do what a pilot does in a flight deck. It can do what ground crews do, and it can deliver a package,” Mr Carbon says.

That autonomy is not absolute. The Darlington flights are conducted “beyond visual line of sight”, BVLOS in industry parlance, but every aircraft is monitored remotely by an operator who liaises with air traffic control at nearby Teesside Airport when required.

The choice of Darlington is, on closer inspection, a piece of careful corporate scouting rather than an accident of geography. The town offers a useful mix of residential streets, major roads and an airport in close proximity, allowing Amazon to stress-test its kit across multiple environments without travelling far. Crucially, it sits beside an Amazon hub with the deep stock needed to support the service.

It is also the only location outside the United States where the company is operating drone deliveries.

The Civil Aviation Authority has granted approval for a trial running to the end of the year, with temporary protected airspace, a regulatory prerequisite for autonomous flight under current rules, secured until mid-June and expected to be extended. Darlington Borough Council, which approved temporary planning permission for what it described as the “unprecedented nature of the scheme”, said it was “great to see Darlington at the forefront of such a pioneering scheme which highlights our borough as an area of innovation, development and investment”.

The limits of flying logistics

For all the choreography, the technology has obvious constraints. Eligible customers will need a garden or yard. Flats and terraces without outside space are excluded.

Dr Anna Jackman, an associate professor of geography at the University of Reading, says the Darlington trial illustrates both the promise and the limitations of the technology. “A lot of our demand for delivery services is in urban centres. They are very densely populated, very congested. And the reality is [drone deliveries] don’t work well in high-rise buildings.”

Rooftop drop-offs and centrally located drone hubs are being explored, she adds, “but right now we’re not there yet”.

There is also the question of safety, where Amazon’s record is not unblemished. In February, an MK30 drone clipped the gutter of an apartment building in a Dallas suburb after losing GPS signal, falling to the ground and breaking apart. No one was hurt, and Amazon has since suspended deliveries to similar buildings. Mr Carbon describes it as one of the “things we learn as we go along”, noting that 170,000 drone flights have been completed safely.

Drones are not entirely new to British skies. The NHS is trialling them to ferry blood supplies across London, and Royal Mail is using them to reach remote communities in Orkney. Amazon’s intervention is different in character: this is a commercial play by the country’s largest online retailer, and the read-across for smaller businesses is significant.

Independent retailers and the SMEs that use Amazon’s marketplace will, sooner or later, face customers who have come to view sub-two-hour delivery as the baseline. The pressure to match, or at least mitigate, that experience will fall hardest on those without the logistics muscle of a global platform. At the same time, the gradual normalisation of BVLOS flight could open new commercial doors for British drone operators, software firms and aerospace suppliers servicing the sector.

For now, the residents of Darlington are the test market, and reaction has been mixed. The launch itself ran years behind Amazon’s original 2023 pledge to begin in 2024, a reminder that aviation regulation does not bend easily to Silicon Valley timelines.

Mr Carbon is unrepentant. “We wouldn’t be doing it if it wasn’t commercially viable,” he says. “It’s a business, right? Absolutely, it can be commercially viable, and that’s the goal that we’re going after.”

Whether it ends up reshaping British retail logistics or remaining an expensively engineered curiosity will depend on what happens next: the regulator’s willingness to widen the airspace, Amazon’s appetite to keep spending, and customers’ willingness to look up.

Read more:
Amazon’s drones touch down in Darlington in UK delivery first

For many leaders, digital transformation has long been something to tackle when time allowed, after the next funding round, after the next product launch, after the next operational fire was put out.

Marcin Pichur, Docuware, Regional Vice President Sales, UK/IRE, Spain, Italy, Poland, explains that the UK and Ireland now setting firm timelines for mandatory e‑invoicing, the era of “digital later” has officially ended.

In the UK, April 2029 has become the defining milestone for finance and IT teams. In Ireland, the deadlines arrive even sooner. Large organisations must comply by late 2028, and every business -regardless of size – must be capable of receiving structured e‑invoices by November of that year. For businesses, this means the countdown has already begun. Even if you only issue a handful of invoices a month, your systems will still need to handle structured data, not PDFs that merely mimic digital progress.

What’s often overlooked is that this shift is not simply a compliance exercise. This is a rare opportunity to modernise finance operations, eliminate manual friction and build a more resilient, data‑driven business. Those who act early will gain a meaningful operational advantage. Those who wait will find themselves scrambling to comply while competitors quietly accelerate.

Europe has already proven the model works

The UK and Ireland are not stepping into uncharted territory. Across Europe, e‑invoicing has already transformed how businesses operate. Italy’s Sistema di Interscambio (SdI) has shown how real‑time reporting can dramatically reduce VAT fraud while forcing a step‑change in business digitisation. France, Spain and Poland are following suit, each using structured invoicing to modernise B2B trade and improve tax transparency.

The results are consistent: real‑time visibility, fewer errors, faster payments and a more predictable cash‑flow environment. For SMEs, where cash flow is often the difference between growth and survival, this level of visibility is truly nothing short of transformative.

One misconception persists, however – the belief that emailing a PDF is digital enough. A PDF is not an e‑invoice. It is digital paper. It still requires manual keying, error‑prone OCR and endless reconciliation work. True e‑invoicing uses structured data (typically XML following the EN 16931 standard) that flows directly from one system to another without human intervention. This is the leap UK and Irish businesses must prepare for, and one that exposes the fragility of many finance processes.

IDP: the missing link that makes e‑invoicing viable

This is precisely where Intelligent Document Processing (IDP) becomes indispensable. If e‑invoicing is the destination, IDP is the engine that can get you there without chaos.

Most SMEs do not operate with pristine data, perfectly aligned supplier records or a single unified ERP. They operate with a blend of accounting tools, spreadsheets, legacy systems and manual workarounds. IDP provides the orchestration layer that makes structured invoicing viable in the real world, not just in policy documents.

Modern IDP platforms can extract, validate and match data across the likes of invoices, purchase orders, goods‑received notes and statements. They can identify discrepancies before they become problems, flag exceptions automatically and create touchless workflows that eliminate manual checking. Crucially, IDP validates data before an invoice leaves your system, ensuring that VAT numbers, line items and PO references are correct. This prevents the rejection loops that drain resources and delay payments, a hidden cost that many underestimate until it becomes a crisis.

For businesses with lean finance teams, IDP is a realistic way to scale without adding headcount. It protects your business from the administrative burden of compliance while laying the foundations for automation that goes far beyond invoicing.

Avoiding the “integration tax”

The challenge for many SMEs is what some call the “integration tax”. Large enterprises have transformation budgets and IT teams. Start‑ups have agility. SMEs often have neither. They are caught between ambition and legacy systems, between the desire to modernise and the reality of limited resources.

Waiting until 2028 or 2029 will only make this worse. A last‑minute scramble leads to rushed implementations, bolt‑on tools that don’t integrate and processes that meet the mandate but do nothing to improve the business. Early adopters, on the other hand, can use the mandate as a driving force to fix long‑standing inefficiencies. They can clean supplier data, eliminate spreadsheet‑driven processes, standardise approvals and build a finance stack that supports growth rather than constraining it. This is where SMEs can turn compliance into a competitive advantage, by treating the mandate not as an obligation but as an opportunity.

For SMEs trading across borders, the complexity increases. Each country has its own tax authority, schema updates and technical requirements. Trying to manage this with multiple tools creates inconsistency and unnecessary risk. The smarter approach is to adopt a single e‑invoicing gateway that manages multi‑country compliance and shields core systems from constant regulatory change, giving businesses the stability to focus on growth rather than chasing tax updates – an outcome an e‑invoicing service like DocuWare’s is designed to deliver.

E‑invoicing is only the beginning

Once structured invoice data flows into your business in real time, the benefits extend far beyond compliance. Cash‑flow forecasting becomes more accurate. Month‑end closes become faster. Supplier relationships improve. Audit trails strengthen. Financial reporting becomes more reliable. And, perhaps most importantly, you can gain the data foundation required for AI‑driven analytics and automation. Finance shifts from a reactive function to a strategic one.

For UK and Irish SMEs, the e‑invoicing mandate is a once‑in‑a‑generation chance to modernise. Those who start now will reduce manual workload, improve cash flow, strengthen compliance and build scalable finance operations long before the mandate arrives. Those who wait will face a rushed, expensive, compliance‑only project that delivers none of the upside.

The shift from digital paper to structured data is already underway. The only decision left is whether your business uses this moment to get ahead or simply to catch up.

At DocuWare, we anticipate regulatory shifts long before they become urgent, giving you the power to act while others scramble. Speak with our experts today and claim your competitive advantage.

Read more:
E-invoicing: A mandate that marks the end of “digital later”

Britain’s sole traders and small business owners can now generate an indicative insurance quote without ever leaving ChatGPT, after digital broker Simply Business became the first in the UK to plug its pricing engine directly into OpenAI’s chatbot.

The London-headquartered insurer, which counts more than one million customers across the UK and United States, has switched on a dedicated app inside ChatGPT’s App Directory. A parallel launch has gone live in the US market on the same day.

For the estimated 5.5 million small businesses across the UK, the pitch is one of speed. Users are asked for just four details , their trade, annual turnover, years trading and UK postcode – and the app returns an indicative price in seconds. Those who wish to proceed are routed to the Simply Business website to complete underwriting and purchase a policy in the conventional way.

The company says the integration has been built with the privacy, security and reliability safeguards that brokers are expected to uphold, a point likely to matter to regulators watching the rapid encroachment of generative AI into regulated financial services.

The move is the latest plank in a global technology strategy that has been gathering pace at Simply Business. In October last year, the firm rolled out a hyper-personalised AI advisor in the US, designed to strip friction out of a purchase journey that has long been a source of frustration for time-poor entrepreneurs.

Group chief executive David Summers said the launch was a natural extension of the company’s founding ambition. “In 2005, we set out to change the way small businesses purchase insurance,” he said. “More than two decades later, we have over one million customers worldwide and we are continuing to evolve our capabilities to simplify the way they research and buy insurance. Launching this insurance app in the UK and the US for small businesses in ChatGPT is our latest step in meeting our customers where they are and making the insurance-buying process an easier, better and fairer experience for them.”

Group chief technology officer Dana Edwards argued that the broker was simply following its customers. “Small business owners are already using platforms like ChatGPT to research, plan and make decisions,” she said. “By safely bringing insurance pricing into that environment, we’re removing one more barrier between them and the coverage they need. We designed the app with the safeguards that customers have come to expect, this kind of rapid, responsible innovation is precisely what our global technology platform is built for.”

The launch underscores a broader shift in how UK SMEs are expected to transact with financial services providers. As conversational AI becomes the first port of call for research on everything from tax to staffing, insurers, accountants and lenders are under growing pressure to meet customers inside those platforms rather than waiting for them to arrive on a branded website.

The Simply Business app will appear as a recommendation when ChatGPT users ask questions related to business risk and insurance cover, or it can be summoned directly from the App Directory.

Read more:
Simply Business becomes first UK broker to put small business cover inside ChatGPT

After 15 transformative years at the helm of the world’s most valuable company, Tim Cook is stepping aside as chief executive of Apple, with hardware engineering chief John Ternus set to inherit one of the most coveted seats in global business.

The Cupertino-based group confirmed on Monday that Cook, 65, will become executive chairman of the board on 1 September, with Ternus, senior vice president of hardware engineering, promoted to chief executive on the same date. The succession, approved unanimously by directors, caps what insiders describe as a patient, long-planned handover rather than a hurried passing of the baton.

Cook will remain chief executive through the summer, working alongside his successor to ensure a seamless transition. In his new chairman’s role, he is expected to focus on global policy engagement, a brief that has grown increasingly weighty as Apple navigates tariff regimes, artificial intelligence regulation and geopolitical pressure on its supply chain.

“It has been the greatest privilege of my life to be the CEO of Apple,” Cook said in a statement. “John Ternus has the mind of an engineer, the soul of an innovator, and the heart to lead with integrity and with honour. He is without question the right person to lead Apple into the future.”

The numbers behind Cook’s tenure make for arresting reading. Since succeeding the late Steve Jobs in 2011, Apple’s market capitalisation has swelled from roughly $350bn to $4tn, a gain of more than 1,000 per cent. Annual revenue has almost quadrupled, climbing from $108bn in the 2011 financial year to more than $416bn in 2025. Cook has added Apple Watch, AirPods and Vision Pro to the firm’s hardware roster, while the Services division he championed now generates more than $100bn a year,  a standalone business that would rank inside the Fortune 40.

For British SMEs that built livelihoods around Apple’s ecosystem, from App Store developers in Shoreditch to hardware resellers on the high street, Cook’s legacy has been the steady expansion of a platform that now reaches 2.5 billion active devices across more than 200 countries. Apple’s global retail footprint has more than doubled during his reign.

Ternus, who has spent almost a quarter of a century at the company, represents a return to the engineer-led tradition established by Jobs. He joined Apple’s product design team in 2001, rose to vice president of hardware engineering in 2013 and entered the executive suite in 2021. His fingerprints are on every major product line, from iPad and AirPods to the recent MacBook Neo and the iPhone 17 range, including the ultra-slim iPhone Air that launched last autumn.

“I am profoundly grateful for this opportunity to carry Apple’s mission forward,” Ternus said. “Having spent almost my entire career at Apple, I have been lucky to have worked under Steve Jobs and to have had Tim Cook as my mentor.”

A Mechanical Engineering graduate of the University of Pennsylvania, Ternus cut his teeth at Virtual Research Systems before joining Apple. He has overseen the transition to Apple-designed silicon, the push into recycled aluminium and 3D-printed titanium, and the evolution of AirPods into an over-the-counter hearing aid, a rare example of Big Tech hardware being cleared as a bona fide medical device.

In a further reshuffle, Arthur Levinson, Apple’s non-executive chairman for the past 15 years, will step back to become lead independent director when the new regime takes effect. Ternus will join the board the same day.

“Tim’s unprecedented and outstanding leadership has transformed Apple into the world’s best company,” said Levinson. “We believe John is the best possible leader to succeed Tim.”

Cook’s departure from the chief executive’s office closes a chapter defined as much by stewardship as by showmanship. Where Jobs dazzled, Cook disciplined — turning a maverick product house into an operational juggernaut, reducing Apple’s carbon footprint by more than 60 per cent against 2015 levels even as revenue roughly doubled, and placing privacy at the heart of the brand proposition. Whether Ternus can continue that trajectory while reigniting the pace of hardware breakthrough will define the next era in Cupertino, and reverberate through every business, large and small, that lives within Apple’s orbit.

Read more:
Cook hands Apple’s reins to Ternus as engineering chief prepares for top job

Mark Zuckerberg is preparing to take the knife to his own creation once again.

Meta Platforms, the parent of Facebook, Instagram and WhatsApp, is lining up a global redundancy programme that will see roughly one in ten of its staff, about 8,000 people, shown the door from next month, with a second wave expected before the year is out.

The Silicon Valley giant has declined to put any figures on the record, but the direction of travel will be uncomfortably familiar to the tens of thousands of staff who lived through Meta’s self-styled “year of efficiency” in 2022 and 2023, when some 21,000 roles were stripped out as the share price slid and the company came to terms with a bout of Covid-era over-hiring.

This time round, the rationale is rather different. Meta is in robust financial health, but Mr Zuckerberg has committed to spending hundreds of billions of dollars reshaping the business around artificial intelligence. The trade-off, it seems, is that a leaner organisation with fewer management layers and AI-augmented engineers is expected to do the heavy lifting that armies of human employees once did.

According to Reuters, the initial tranche of cuts is pencilled in for May, with the timing and scope of the later round yet to be nailed down. Meta employed just shy of 79,000 people at the end of December, according to its most recent filing, meaning the opening salvo alone could remove close to a tenth of that headcount.

Meta is not moving in isolation. Amazon has already swept out 30,000 corporate staff in recent months, equivalent to nearly ten per cent of its white-collar base, while in February the fintech group Block let go of nearly half its workforce, around 4,000 jobs. In both cases, senior management pointed firmly at efficiency gains from AI as the justification.

The industry’s own body count bears that out. Layoffs.fyi, which tracks redundancies across the technology sector, puts the tally at 73,212 jobs lost in the first four months of 2026 alone. For the whole of 2024, the figure was 153,000, suggesting this year’s numbers are on course to eclipse anything seen in the post-pandemic shake-out.

Inside Meta, the reorganisation is already well under way. Teams within its Reality Labs division have been reshuffled in recent weeks, and engineers from across the group have been parachuted into a newly minted Applied AI unit. Its brief is to accelerate the development of AI agents capable of writing code and executing complex tasks without human hand-holding, the very capability, critics will note, that Mr Zuckerberg appears to believe can replace a sizeable chunk of his own workforce.

For Britain’s small and medium-sized businesses watching from across the Atlantic, the signal is a telling one. When the world’s largest technology employers openly argue that generative AI is now capable enough to displace thousands of skilled knowledge workers, the pressure on every other business to rethink how it organises, recruits and deploys talent only intensifies.

Whether the efficiency dividend materialises as cleanly as Mr Zuckerberg hopes remains to be seen. Meta’s 2022 cuts were followed by a sharp recovery in profitability and a soaring share price, vindicating his tough love approach in the eyes of Wall Street. A second act on a similar scale, however, will test whether AI can genuinely deliver the productivity miracle its champions promise, or whether Meta is simply exchanging one kind of risk for another.

Read more:
Meta to axe 8,000 jobs in May as Zuckerberg bets the house on AI

A powerful new artificial intelligence model developed by Anthropic has triggered a flurry of crisis meetings among finance ministers, central bankers and senior financiers, who fear the technology could be turned on the global financial system with devastating consequences.

The model, known as Claude Mythos, has been shown to pinpoint vulnerabilities in many of the world’s most widely used operating systems, prompting alarm at the highest levels of government and commerce. While some specialists believe it marks a step-change in AI’s ability to uncover and exploit cyber-security flaws, others have urged caution, arguing that far more independent testing is needed before its true capabilities can be judged.

Canada’s Finance Minister, François-Philippe Champagne, confirmed to media that Mythos had dominated discussions at this week’s International Monetary Fund meetings in Washington DC. “Certainly it is serious enough to warrant the attention of all the finance ministers,” he said. Drawing a comparison with geopolitical risks, he added: “The difference is that the Strait of Hormuz – we know where it is and we know how large it is… the issue that we’re facing with Anthropic is that it’s the unknown, unknown. This is requiring a lot of attention so that we have safeguards, and we have processes in place to make sure that we ensure the resiliency of our financial systems.”

Mythos is among the latest additions to Anthropic’s Claude family of models, which competes directly with OpenAI’s ChatGPT and Google’s Gemini. It was unveiled earlier this month by developers responsible for stress-testing so-called “misaligned” AI behaviour, instances in which a model acts against human values or intended goals. Their verdict was that Mythos is “strikingly capable at computer security tasks”.

Citing concerns that the model could surface long-dormant software bugs or identify novel ways to exploit system weaknesses, Anthropic has opted not to release it publicly. Instead, access has been granted to a handful of technology giants, including Amazon Web Services, CrowdStrike, Microsoft and Nvidia, under an initiative dubbed Project Glasswing, which the company describes as an “effort to secure the world’s most critical software”.

On Thursday, Anthropic released an upgraded version of its existing Claude Opus model, saying this would enable Mythos’s cyber capabilities to be evaluated within less powerful systems.

Not everyone in the cyber-security community is convinced the fears are proportionate, particularly given the limited independent testing conducted so far. The UK’s AI Security Institute, which has been given access to a preview version, is the only body to have published an independent assessment. Its researchers concluded that while Mythos Preview could compromise systems with weak defences, it was not dramatically more capable than its predecessor, Opus 4. “Our testing shows that Mythos Preview can exploit systems with weak security posture, and it is likely that more models with these capabilities will be developed,” the report’s authors wrote.

Sceptics have also pointed to precedent: in February 2019, OpenAI similarly delayed the release of GPT-2 on safety grounds, a decision critics at the time dismissed as a marketing device.

Senior bankers are now to be granted early access to Mythos so they can probe their own defences ahead of any wider release. C.S. Venkatakrishnan, chief executive of Barclays, told the BBC: “It’s serious enough that people have to worry. We have to understand it better, and we have to understand the vulnerabilities that are being exposed and fix them quickly.” He added that a far more interconnected financial system had created both fresh opportunities and fresh exposures, cautioning: “This is what the new world is going to be.”

For Britain’s small and medium-sized businesses, which rely on the integrity of banking, payment and cloud infrastructure every day, the implications are considerable. A cyber incident capable of destabilising a major lender or payment processor could ripple rapidly through SME supply chains, hitting cash flow, invoicing and customer confidence within hours.

Anthropic has already flagged that Mythos has uncovered multiple vulnerabilities in core operating systems, financial platforms and web browsers. Governments and banks are being offered advance access to harden their defences before any public launch.

Andrew Bailey, Governor of the Bank of England, has said that the development must be treated with the utmost seriousness. “We are having to look very carefully now what this latest AI development could mean for the risk of cyber crime,” he said. “The consequence could be that there is a development of AI, of modelling, which makes it easier to detect existing vulnerabilities in sort of core IT systems, and then obviously cyber criminals, the bad actors, could seek to exploit them.”

The US Treasury has confirmed that it has raised the matter directly with major American banks, urging them to run internal tests ahead of any public release. Industry sources further suggest that a rival US AI firm could shortly unveil a similarly potent model, but without comparable guardrails.

For the UK technology sector, the controversy may prove an opening as much as a threat. James Wise, a partner at Balderton Capital and chair of the newly established Sovereign AI unit, a £500m government-backed venture capital fund targeting home-grown AI businesses, argued that Mythos is merely “the first of what will be many more powerful models” capable of exposing systemic weaknesses.

Speaking to the BBC’s Today programme, he said his unit was “investing in British AI companies that are tackling that, companies working in AI security and safety”, adding: “We hope the models that expose vulnerabilities are also the models which will fix them.”

For the country’s AI scale-ups and cyber-security start-ups, the message from Threadneedle Street and Washington alike is unmistakable: the defensive side of the AI arms race has just become one of the most commercially significant frontiers in British enterprise.

Read more:
Finance chiefs sound alarm over Anthropic’s ‘mythos’ AI model amid cyber-security fears

Britain’s video games industry is at risk of haemorrhaging talent and intellectual property to more nimble overseas rivals unless Whitehall moves swiftly to sharpen its tax and investment incentives, a leading advisory firm has warned.

With France, Ireland and Australia aggressively courting studios through increasingly generous reliefs, the UK’s reputation as a global gaming powerhouse, home to franchises from Grand Theft Auto to Tomb Raider, could begin to slip, according to audit, tax and business advisory firm Blick Rothenberg.

Speaking during London Games Festival week, Mandy Girder, a partner at the firm, said the sector urgently needed the Government to “level up” its support if Britain was to keep its seat at the top table of global games development.

“Without decisive action from the Government, the UK risks losing both talent and intellectual property to other countries,” she said. “France, Australia and Ireland are offering increasingly generous and accessible incentive regimes designed to attract investment.”

The London Games Festival, now a fixture in the industry calendar, has put a spotlight on British creativity, but Girder cautioned that creativity alone would not keep the UK ahead of the pack.

“The festival highlights the UK’s undeniable creative strength, but creativity alone will not secure long-term global leadership,” she said. “The Government must step up tax relief and investment in the industry.”

While the UK’s Video Games Expenditure Credit and broader creative industry reliefs have underpinned growth in recent years, Girder warned that the regime was increasingly seen by studios as cumbersome when set beside rivals abroad.

“Headline rates are competitive, but the system is often viewed as more complex and, in some cases, less flexible or accessible than the incentive regimes in countries such as Ireland and Australia,” she said.

Recent tightening of eligibility rules is already beginning to bite. Under the revised framework, at least 10 per cent of development costs must now be incurred in the UK rather than across the wider European Economic Area, a change intended to bolster domestic employment but which has tripped up projects structured around continental teams.

“Whilst intended to encourage the use of UK-based talent, this has been restrictive on the number of successful claims for projects already under way and structured around European teams,” Girder said. “It has led to a decline in the availability of these tax credits.”

She is calling for a simpler, more generous regime, backed by targeted incentives explicitly designed to draw inward investment.

“Simplifying and enhancing the UK’s tax framework, alongside introducing targeted incentives to attract inward investment, would significantly strengthen the UK’s global positioning,” she said.

Access to finance is another persistent headache, particularly for studios trying to move beyond the start-up phase. While seed capital is relatively easy to come by, scale-up funding, the kind that allows mid-sized studios to expand internationally and retain their IP, remains elusive.

“Early-stage funding is relatively accessible, but mid-sized studios often face barriers when seeking the scale-up capital needed to expand internationally and retain valuable intellectual property,” Girder said. “This funding gap risks limiting the UK’s ability to fully capitalise on its creative strengths.”

The Government’s newly launched Creative Industries Sector Plan, which opens £28.5 million in funding for the next generation of games developers, is a step in the right direction, Girder conceded.

“The UK has long been recognised as a creative powerhouse, home to world-class studios and exceptional talent behind globally successful titles such as Grand Theft Auto and Tomb Raider,” she said. “The sector plan is a positive step forward.”

But she questioned whether the intervention goes far enough to tackle the structural weaknesses in the industry’s funding pipeline.

“The question remains whether this level of support is sufficient to address the structural funding challenges facing the sector,” she said. “A more comprehensive approach, combining competitive tax relief, grants and alternative financing options, will be essential to unlock sustained growth.”

Her message to ministers was blunt. “Now is the time for industry and Government to work together to simplify incentives, unlock scale-up funding, and ensure the UK remains a destination of choice for global games investment.

“The London Games Festival turns the spotlight on the UK’s role as a leading force in the global video games market, and on the steps the Government needs to take to secure its future competitiveness.”

Read more:
Britain’s gaming industry needs a power-up or risks losing its crown to France, Ireland and Australia

The government has fired the starting gun on a £30 million funding offensive aimed at Britain’s video games sector, urging developers with ambitions to create the next blockbuster title to come forward for a share of the pot.

At the heart of the package is a £28.5 million UK Games Fund, which effectively doubles previous public investment in the industry. Applications open from 14 April, with grants split across three tiers designed to support studios at every stage of growth, from fledgling outfits with little more than a strong concept through to established developers ready to take a game to market.

The entry track offers up to £20,000 for newly formed companies showing genuine potential. An emergent track provides up to £100,000 for prototyping, while the expansion track, carrying grants of up to £250,000, the largest the fund has ever offered, is aimed at studios looking to complete a title and scale their operations.

The remaining £1.5 million has been earmarked for the London Games Festival over the next three years, with the stated aim of strengthening investor partnerships and doubling the value of private investment deals brokered at the event to £30 million annually.

Creative Industries Minister Ian Murray was characteristically blunt about the rationale. The gaming sector, he argued, has been undervalued for too long despite its considerable economic heft. Britain’s gaming market now generates £8.8 billion in consumer spending each year, and the country is home to more than 2,000 games companies whose output, from Grand Theft Auto and Tomb Raider to PowerWash Simulator and No Man’s Sky, has defined genres and built global audiences.

For small and medium-sized studios, the funding architecture matters as much as the headline figure. Access to finance has long been the sector’s most persistent constraint, particularly for independents operating outside the orbit of major publishers. Dr Richard Wilson OBE, chief executive of trade body TIGA, noted that the organisation has repeatedly called for greater prototype and content funding to help studios bridge the gap between concept and commercial product.

The geographical spread of the industry adds a further dimension. While London remains a significant hub, gaming has deep roots in Dundee, Leamington Spa and Guildford, among other locations. The Tay Cities Region has already received £20 million in government backing to advance creative technologies including games and virtual reality, a signal that ministers see the sector as a genuine vehicle for regional economic development rather than a metropolitan concern.

The Games Growth Package forms a central plank of the Industrial Strategy’s Creative Industries Sector Plan, a £380 million blueprint published earlier this year. It sits alongside enhanced support from the British Business Bank, UK Research and Innovation, and the existing games tax relief regime, which has been one of the more effective fiscal interventions in the creative industries since its introduction.

The package has drawn a broadly positive response from across the industry. Nick Poole OBE, chief executive of Ukie, described it as a strong vote of confidence in British gaming, while Nick Button-Brown, chair of the UK Video Games Council, called it an “amazing statement of intent” regarding long-term government commitment.

Beyond funding, the government is also turning its attention to the consumer side of the market. The Chartered Trading Standards Institute has been commissioned to develop guidance clarifying the rights of gamers purchasing digital content, with a consultation expected in the coming months. Ministers will also engage with the newly established UK Esports Advisory Panel, a Ukie-led forum intended to keep Britain competitive in the rapidly expanding competitive gaming space.

For the thousands of small studios and independent developers that form the backbone of the British games industry, the practical question now is whether the money can flow quickly enough and flexibly enough to make a material difference. The three-tier structure suggests the government has at least listened to the sector’s concerns about accessibility. Whether it proves sufficient to keep British talent from being lured abroad by better-funded competitors remains to be seen.

Read more:
Government doubles down on gaming with £30m funding package as sector eyes global growth

British businesses are in danger of being left stranded in the middle of the pack on artificial intelligence, with a new PwC study revealing a significant gap between UK firms and the world’s top AI adopters in both spending and returns.

The consultancy’s global survey found that while leading companies worldwide are investing an average of five per cent of revenue in AI and reaping returns of 15 per cent, their British counterparts are committing just two per cent and generating returns of ten per cent. It is a gap that should alarm boardrooms across the country, particularly among small and medium-sized enterprises already grappling with tight margins and limited budgets for technology transformation.

Perhaps more troubling is the innovation shortfall the figures suggest. UK businesses derived only 27 per cent of their revenue from products that did not exist three years ago, compared with 43 per cent among global leaders. For SMEs, which have historically relied on agility and fresh thinking to compete against larger rivals, that disparity ought to prompt some uncomfortable questions about whether enough is being done to turn AI capability into genuinely new commercial offerings.

The research points to familiar obstacles. Outdated IT systems and rigid internal processes continue to hold companies back, with only 27 per cent of UK businesses having redesigned their workflows to properly integrate AI rather than simply grafting it on to what already exists. The same proportion had modernised legacy technology to better accommodate the tools.

There is also a question of ambition. Nearly half of the UK businesses surveyed said efficiency and productivity were their primary motivation for experimenting with AI, while just 26 per cent cited revenue generation. It is a mindset that Leigh Bates, PwC UK’s global risk AI leader, believes is limiting the country’s potential.

Bates described the findings as a wake-up call, arguing that too many firms remain trapped between piloting AI projects and scaling them effectively. The businesses seeing the greatest returns globally, he said, are not merely doing more of the same but fundamentally reinventing how they operate.

Overall, the UK ranked 11th out of 19 countries in PwC’s assessment, behind China at the top of the table, as well as France, Germany and Saudi Arabia. The United States, notably, fared little better at 13th. PwC defined global leaders as companies in the top 20 per cent of AI-driven performance.

For Britain’s SME community, the message is clear enough. The window to move from cautious experimentation to meaningful adoption is narrowing, and those that continue to treat AI as little more than a cost-cutting exercise risk discovering that their competitors, at home and abroad, have already moved on.

Read more:
UK firms risk being left behind as AI adoption gap widens, warns PwC

Google has fired the opening shot in a battle for Britain’s restaurant booking market, rolling out an artificial intelligence tool that allows diners to secure a table without ever leaving the search bar.

The feature, which went live on Friday, invites users to describe the sort of meal they are after in plain language. Google’s AI then trawls the web for real-time availability and returns a shortlist of bookable options within seconds, collapsing what was once a multi-step hunt into a single query.

It represents a marked departure from the traditional search experience. Rather than directing punters off to comparison sites or third-party platforms, Google is now intent on keeping the entire customer journey, from the first idle thought about dinner through to a confirmed reservation, firmly within its own walls.

The Silicon Valley giant said appetite for smarter dining tools is growing sharply, pointing to a 140 per cent rise this year in search queries such as “when to book a table” as consumers demand faster and more tailored recommendations.

Listings will be drawn from partners including TheFork, Sevenrooms and DesignMyNight, yet the interface, and crucially the customer relationship, will sit squarely with Google. That raises awkward questions about who ultimately owns the diner and who profits from the transaction.

The move sets Google on a direct collision course with established players such as OpenTable, whose business has long depended on playing intermediary between restaurants and hungry customers. By intercepting users at the point of search and ushering them through to booking, Google threatens to disintermediate those platforms altogether and squeeze their margins in the process.

More broadly, the launch signals the dawn of a new phase in the AI race, one defined not by chatbots answering questions but by agents quietly completing tasks on the user’s behalf. The ultimate prize is a search engine that functions as a digital concierge, and for Google, controlling bookings delivers a rich seam of behavioural data that can be fed back into its advertising and recommendation machinery.

Britain, with its densely packed restaurant scene and enthusiastic take-up of online reservation platforms, offers an ideal proving ground before the technology is extended to adjacent sectors such as travel and live events. For the incumbents of the booking world, the writing may already be on the wall.

A spokesman for Opentable said: “OpenTable is a partner in this launch with Google, and our inclusion is the latest example of how we’re meeting diners wherever they are, and on whatever platform they’re using. This expansion with Google builds on our existing partnerships with OpenAI, Perplexity, and Alexa+ to make it easier than ever for diners to find and secure the right table in the moment.

“For our restaurant partners, we continue to support them in finding new ways to fill seats and maximise revenue by surfacing real-time availability at the point of search. OpenTable continues to be a key connector between restaurants and diners, and this integration is no different.”

Read more:
Google takes on Opentable with AI that books your dinner in seconds

Kate Alessi, Google’s managing director for the UK and Ireland, has pushed back firmly against warnings that artificial intelligence will trigger widespread unemployment, insisting that the greater risk lies in failing to equip workers with the skills to thrive alongside the technology.

Speaking as Google unveiled a new national upskilling programme backed by £2 million in grant funding from Google.org, Alessi argued that history offered a reassuring precedent. Every previous wave of technological disruption, she noted, had prompted the same anxieties about disappearing jobs – and every time, the fears had proved overblown as new roles emerged to replace the old.

Her intervention comes at a pointed moment. In January, the Mayor of London, Sadiq Khan, cautioned that AI could bring about a new era of mass unemployment without proper oversight, while Bank of England governor Andrew Bailey drew comparisons with the Industrial Revolution, stressing the need for retraining and education on a significant scale.

Alessi does not deny that change is coming, but she frames it rather differently. Citing research from the policy consultancy Public First, she pointed out that roughly six in ten UK jobs are expected to be enhanced rather than eliminated by AI. The challenge, she maintained, is ensuring that people are prepared to step into the roles the technology creates, not simply bracing for the ones it displaces.

The figures suggest there is considerable ground to make up. According to new research commissioned by Google, although nearly two thirds of the UK population have tried AI tools, just one in ten consider themselves advanced users. Only a quarter felt they were deploying AI in ways that saved them meaningful time or gave them genuinely new capabilities.

“Most people are really only scratching the surface,” Alessi said.

To address that gap, Google is rolling out a series of practical initiatives. Alongside the grant funding, the company plans to run Gemini tours across universities, aimed at ensuring graduates enter the workplace with a working knowledge of AI. It will also stage a series of pop-up events branded as “squeeze the juice” bars in towns and cities around the country, designed to show ordinary users how to move beyond basic prompting to tackle more complex tasks – from automating routine admin to conducting in-depth research.

Read more:
Don’t fear AI job losses – invest in training, urges Google’s UK boss

Reform UK is venturing into podcasting with a weekly show that will offer listeners behind-the-scenes access to Nigel Farage and senior figures within the party, marking the first time a British political party has produced its own audio programme.

The first episode, due out on Saturday, will feature footage from Reform’s campaign trail ahead of the local elections, including exchanges with both supporters and detractors. Subsequent instalments will follow Farage’s campaigning efforts in Wales and Scotland while covering major policy announcements in depth. The show will be available on Spotify and Apple, though the party has confirmed there are no plans to appoint a regular presenter.

The move represents a significant escalation in Reform’s broader digital media strategy, which has already seen the party invest tens of thousands of pounds in an in-house television studio. Farage commands a social media following of nearly 7.3 million across X, Facebook, TikTok, Instagram and YouTube, a figure that exceeds the combined followings of Sir Keir Starmer, Kemi Badenoch, Sir Ed Davey and Green Party leader Zack Polanski.

That digital dominance has translated into tangible political momentum. Reform now leads the national polls and has become the most popular party among Generation Z men, according to research by JL Partners for the think tank Onward. The party’s sharp use of TikTok has been widely credited as a driving force behind its surge in support among younger voters.

The podcast launch also underscores a growing tension between political parties and traditional broadcast media. Farage already hosts a primetime programme on GB News, a channel that has faced repeated scrutiny from Ofcom over its use of politicians as presenters. Culture Secretary Lisa Nandy has argued that Farage’s show is undermining public trust in news broadcasting.

Reform’s digital success has not gone unnoticed by its rivals. The Prime Minister joined both TikTok and Substack late last year, while Labour has enlisted FourOneOne, a digital marketing agency backed by Silicon Valley investors including LinkedIn founder Reid Hoffman, to mount a campaign targeting Reform on TikTok. The party has further strengthened its online presence following Robert Jenrick’s defection from the Conservatives, with the former shadow justice secretary having built a considerable profile through attention-grabbing social media content.

Farage said the podcast would bring listeners closer to the party’s operations in a way that no other political organisation has attempted, describing it as offering access to every aspect of Reform’s activities.

Read more:
Reform UK becomes first British political party to launch its own podcast

OpenAI’s much-trumpeted plans to build a major data centre in the north-east of England have ground to a halt, dealing a significant blow to Sir Keir Starmer’s strategy of placing artificial intelligence at the centre of Britain’s economic growth.

The maker of ChatGPT announced last September that it would bring its Stargate programme, a global data centre initiative originally valued at $500bn (£378bn), to British shores through a partnership with Nscale, the UK-based data centre operator. The initial plan envisaged housing approximately 8,000 Nvidia AI processors at Cobalt Park on Tyneside during the first quarter of 2026. That deadline has now passed without a spade in the ground, and OpenAI has declined to offer a revised timetable.

The reasons behind the delay remain unclear, though commercial negotiations between the parties are understood to be continuing. Both OpenAI and Nscale refused to comment on the state of play.

The Stargate concept was first unveiled by Sam Altman, OpenAI’s chief executive, at a White House press conference in January 2025 alongside Donald Trump. Altman subsequently pledged to extend the programme internationally, with the UK earmarked as a key location. In a government press release at the time, he described Stargate UK as part of a “shared vision” to expand opportunity through the right infrastructure.

The project was enthusiastically embraced by ministers, who have sought to position Britain as a global leader in AI. OpenAI further signalled its commitment to the UK by appointing George Osborne, the former Conservative chancellor, to spearhead its international expansion.

Yet the Tyneside setback is far from an isolated case. In the United States, negotiations over Stargate’s broader rollout have proceeded sluggishly, with key backer SoftBank among those yet to finalise terms. A planned expansion of a major site in Texas, being developed with the American data giant Oracle, was quietly shelved earlier this year.

The wider industry is grappling with similar headaches. Technology groups have collectively committed to spending hundreds of billions of dollars on data centres to satisfy surging demand for AI applications, but delivery is proving far harder than the headline figures suggest. Research by Sightline Climate indicates that up to half of all large-scale data centre projects are now running behind schedule, hampered by planning difficulties and constraints on energy supply.

Nscale, valued at $15bn and counting Sir Nick Clegg, the former deputy prime minister, among its board members, has itself been forced to push back timelines on a separate development in Loughton, Essex, as Business Matters reported last week.

Critics have been quick to seize on the lack of progress. Tom Hegarty, a spokesman for Foxglove, the campaign group that has raised concerns about the environmental impact of the data centre boom, said the Stargate UK project amounts to little more than a press release issued eight months ago.

The government maintained that it remains focused on fostering the right conditions for investment. A spokesman said ministers are continuing to work with OpenAI and other leading AI firms to strengthen the UK’s computing capacity. Whether that reassurance will be enough to quieten growing scepticism about the pace of delivery is another matter entirely.

Read more:
OpenAI’s flagship UK data centre hits the buffers in blow to Starmer’s AI ambitions

Oracle has begun cutting thousands of jobs as it accelerates a costly push into artificial intelligence infrastructure, with analysts warning the layoffs could ultimately reach tens of thousands of roles.

Employees were informed via email that their positions were being eliminated “as part of a broader organisational change”, with some workers immediately locked out of company systems. The abrupt nature of the cuts has drawn attention across the tech sector, particularly as Oracle seeks to free up capital for its expanding AI ambitions.

The company, founded by Larry Ellison, employs around 160,000 people globally, and analysts have suggested that between 20,000 and 30,000 jobs could be at risk as part of the restructuring.

The layoffs come amid a major shift in Oracle’s strategy, as it commits tens of billions of dollars to building data centres to support the rapid growth of artificial intelligence.

The company has forecast spending of up to $50 billion this year alone on new infrastructure, designed to provide computing power for major clients including OpenAI and Meta.

This follows a landmark agreement with OpenAI, which said it would spend around $300 billion over time on AI processing capacity, a deal that initially boosted investor confidence but has since raised concerns about execution risk and financial exposure.

Oracle’s share price has fallen sharply in recent months, shedding around half its value as investors question the scale and sustainability of its AI investment strategy.

The company is expected to fund much of its expansion through a combination of debt and equity, prompting fears about balance sheet pressure and the potential for overspending in a highly competitive and rapidly evolving market.

Concerns were heightened when Blue Owl Capital withdrew from financing a $10 billion data centre project in Michigan, signalling growing caution among backers.

Those affected by the cuts have begun speaking publicly, emphasising that the layoffs are not linked to individual performance but to broader strategic changes.

Michael Shepherd, an Oracle operations manager, described the move as a “significant reduction in force” impacting “talented and high-performing people”, reflecting the scale and seriousness of the restructuring.

The cuts are expected to focus heavily on operational and support roles, as the company reallocates resources towards high-growth areas such as cloud computing and AI infrastructure.

Ellison, now 81 and still serving as Oracle’s chief technology officer and largest shareholder, remains central to the company’s strategic direction.

His aggressive push into AI reflects a broader race among technology giants to dominate the next phase of computing, but also carries significant financial risk given the scale of required investment.

Beyond Oracle, Ellison has also been involved in other major ventures, including backing large-scale media acquisitions and maintaining close ties with political and business leaders.

Oracle’s move is part of a wider trend across the technology sector, where companies are restructuring workforces to fund AI development and infrastructure.

As demand for computing power surges, firms are increasingly prioritising capital-intensive investments over traditional operational spending, leading to job cuts even among profitable businesses.

The success of Oracle’s strategy will depend on whether its AI investments deliver sustained growth and returns that justify the scale of spending.

In the short term, the layoffs highlight the trade-offs facing technology companies as they navigate a period of rapid transformation.

For employees, the shift underscores the changing nature of work in the digital economy. For investors, it raises questions about how far companies can go in the race for AI dominance without undermining financial stability.

As the industry continues to evolve, Oracle’s high-stakes bet on AI will be closely watched as a bellwether for the broader tech sector.

Read more:
Oracle cuts thousands of jobs as Ellison doubles down on AI investment

YouTube has been ranked the world’s most influential brand, as technology companies continue to dominate global media and public discourse, according to a new report.

The 2026 Brand Influence Rank from Onclusive found that digital-first platforms occupy every position in the global top 10, reflecting their unrivalled ability to shape narratives across both traditional and social media.

Joining YouTube at the top of the rankings are Google, Instagram, Facebook, LinkedIn, Apple, Amazon, Microsoft, TikTok and ChatGPT, underscoring the structural advantage these businesses hold in driving attention at scale.

The report measures influence not by size alone, but by a brand’s ability to generate sustained media coverage, spark conversation and shape public perception globally.

Digital platforms, with their always-on engagement and vast user bases, are uniquely positioned to dominate this landscape. Their central role in communication, content distribution and increasingly artificial intelligence gives them a powerful edge over traditional brands.

Jennifer Roberts, chief marketing officer at Onclusive, said the findings reflect a fundamental shift in how influence is defined.

“Influence is no longer just about reputation, it’s about the ability to generate continuous attention across multiple channels,” she said, noting that the rise of AI-driven search and content is accelerating this trend.

One of the most notable developments in the rankings is the entry of ChatGPT into the global top 10 for the first time, highlighting the rapid ascent of AI-focused brands.

Alongside Microsoft, AI platforms are generating disproportionate levels of media coverage, driven by innovation, competition and ongoing debate around regulation, ethics and the future of work.

However, this visibility comes with a trade-off. The report identifies a “sentiment ceiling” affecting many leading tech brands, where high levels of scrutiny limit positive perception despite strong influence.

Companies such as Google, Facebook, Apple and TikTok all recorded relatively modest positive sentiment scores, reflecting ongoing regulatory pressures, antitrust investigations and concerns over platform governance.

The report also highlights the growing role of corporate leaders in shaping brand narratives.

Elon Musk was ranked the world’s most influential CEO, with a media presence nearly ten times greater than his closest competitor. His influence is driven by his involvement across multiple high-profile companies, including Tesla, SpaceX and the social platform X, combined with a highly visible and often polarising public persona.

Sam Altman ranked second, reflecting the central role of artificial intelligence in global discourse. His prominence has grown rapidly as AI has become a defining topic in business, politics and society.

Other influential leaders include Mark Zuckerberg, Jensen Huang and Tim Cook, each contributing to their companies’ visibility through strategic positioning in key technology sectors.

The report underscores a key tension in modern brand building: influence does not necessarily equate to positive sentiment.

While tech companies dominate attention and conversation, they also face intense scrutiny over issues ranging from data privacy and competition to the societal impact of their technologies.

This dynamic creates a balancing act for brands, which must manage both visibility and trust in an increasingly complex media environment.

As digital platforms and AI continue to reshape how information is created, distributed and consumed, their dominance in global influence rankings is likely to persist.

However, with that influence comes heightened responsibility, and greater scrutiny.

For brands, the challenge is no longer simply to be seen, but to be trusted.

Read more:
YouTube named world’s most influential brand as tech dominance grows

Octopus Investments is set to cut around a fifth of its workforce as it accelerates the adoption of artificial intelligence, in a move that reflects the rapid transformation underway across the asset management industry.

The City-based firm, which manages close to £15 billion in assets, is understood to be placing around 130 roles at risk of redundancy, primarily in back-office functions. With just over 600 employees, the restructuring represents a significant shift in how the business operates, as it seeks to streamline processes and modernise its infrastructure.

The cuts form part of a broader strategy to invest more heavily in technology, particularly AI, which is increasingly being used to automate routine tasks, improve efficiency and reduce operational costs across financial services.

The move underscores how quickly AI is reshaping the financial sector, particularly in areas such as administration, compliance and reporting, where repetitive processes are well suited to automation.

Asset managers have been among the fastest adopters of the technology, using AI tools to handle data processing, client onboarding and portfolio analytics. As a result, roles that were once labour-intensive are being reduced or redefined.

Octopus Investments said the decision was necessary to ensure the business remains competitive in a rapidly changing environment.

“We’ve made the difficult but necessary decision to ensure we are a simpler business that can respond to the pace of change,” a spokesperson said, adding that affected employees would be supported in finding new roles both within the wider group and externally.

The restructuring is not an isolated case. Across the City and globally, financial institutions are reassessing their workforce structures as AI capabilities expand.

HSBC, for example, is reportedly considering up to 20,000 job cuts over the coming years, partly driven by the efficiency gains offered by AI.

The shift reflects a broader recalibration of the industry, where firms are balancing cost pressures with the need to invest in new technologies that can enhance performance and client service.

Despite the job cuts, Octopus Investments remains financially robust. The firm reported a 10.3 per cent increase in net profit to £76.7 million in 2024, with revenues rising to £225.7 million.

It is one of the most profitable divisions within the wider Octopus Group, which also includes businesses such as Octopus Energy and Octopus Money.

The decision to reduce headcount is therefore not driven by financial distress, but by a strategic effort to adapt to technological change and maintain long-term competitiveness.

The firm has faced some criticism in recent years over the fees charged on certain investment products.

Its flagship venture capital trust, Octopus Titan VCT, agreed to reduce management fees by 17 per cent last year, while the company has also earned substantial fees from managing private investment vehicles, even in periods where those funds reported losses.

These issues have added to the pressure on the business to demonstrate efficiency and value for investors, a factor that may also be influencing its push towards automation.

For employees, the restructuring highlights the growing impact of AI on white-collar roles, particularly in financial services.

While front-office and client-facing positions are less immediately affected, back-office functions are increasingly being automated, reducing the need for large operational teams.

At the same time, new roles are emerging in areas such as data science, AI development and digital strategy, suggesting a shift in the types of skills required across the industry.

As AI continues to evolve, asset managers are likely to face further pressure to adapt their business models, balancing efficiency gains with the need to retain expertise and maintain client trust.

For Octopus Investments, the current restructuring represents a significant step in that transition, one that reflects both the opportunities and challenges posed by technological change.

Across the City, similar moves are expected to follow, as firms seek to position themselves for a future where automation plays an increasingly central role in financial decision-making and operations.

Read more:
Octopus Investments to cut 20% of staff as AI reshapes asset management

Oxford spinout Stateful Robotics has raised $4.8 million in pre-seed funding as it looks to solve one of the most persistent challenges in robotics: enabling machines to operate reliably over extended periods in unpredictable real-world environments.

The round was led by Amadeus Capital Partners and Oxford Science Enterprises, with additional backing from serial entrepreneur Stan Boland, founder of autonomous vehicle company Five.

The funding will be used to accelerate deployment of Stateful’s platform, which introduces a new layer of “long-horizon intelligence” — allowing robots to remember past events, adapt to changing conditions and plan tasks over hours or days rather than moments.

While recent advances in large language models and foundation AI systems have significantly improved robots’ ability to perceive and interpret their surroundings, most systems still struggle when environments change.

Unexpected obstacles, shifting lighting conditions or operational disruptions can quickly derail robotic systems that lack the ability to learn from past experiences.

Stateful Robotics aims to address this limitation by building what it describes as a persistent, evolving model of each deployment environment. By continuously integrating data on tasks, performance and historical outcomes, the platform allows robots to anticipate challenges and adapt in real time.

Professor Nick Hawes, co-founder and chief scientist, said traditional systems treat each decision in isolation.

“Stateless systems cannot remember previous incidents or how work actually flows through a site,” he said. “Our platform builds a shared model of tasks and environments that enables robots to adapt to disruption and complete missions safely without constant supervision.”

The company was co-founded by chief executive Kirsty Lloyd-Jukes, previously CEO of Latent Logic, an Oxford spinout acquired by Waymo, alongside leading academic researchers including Professor Nick Hawes, Professor David Parker and Dr Bruno Lacerda.

Their work builds on more than a decade of research at the University of Oxford in areas such as autonomy, decision-making under uncertainty and probabilistic verification.

Lloyd-Jukes said the key challenge facing robotics is not immediate decision-making, but longer-term planning.

“Most robots are good at ‘what now’, but fail at ‘what next’, especially when ‘next’ spans hours or days,” she said. “By maintaining a live model of each deployment, we ensure robots perform reliably and consistently across complex environments.”

Investors believe the technology could help unlock large-scale commercial adoption of robotics across sectors such as logistics, infrastructure, energy and healthcare.

Dr Manjari Chandran-Ramesh of Amadeus Capital said the evolution of robotics, from static industrial arms to mobile systems operating in human environments, requires a new form of intelligence capable of reasoning over time and context.

Similarly, Oxford Science Enterprises highlighted what it sees as a critical bottleneck in the industry: the inability of current systems to handle long-term planning and operational complexity.

Stateful Robotics is already working with pilot customers in sectors including logistics and infrastructure, where reliability and safety are critical to scaling automation.

The new funding will support expansion of its engineering team, further development of its performance engine and broader commercial rollout with industrial partners.

The spinout also reflects the continued strength of the UK’s deep-tech ecosystem, with Oxford University Innovation playing a key role in supporting the company’s formation and early development.

As robotics hardware becomes increasingly mature, attention is shifting to the software and intelligence layers required to make systems truly autonomous.

Stateful Robotics is betting that solving the “memory and planning” problem will be the key to turning promising prototypes into dependable, large-scale solutions, and, in doing so, unlocking the next phase of the automation revolution.

Read more:
Oxford spinout Stateful robotics raises $4.8m to tackle real-world AI for robots

The United States has moved to ban new foreign-made consumer internet routers, citing mounting national security concerns over cyber vulnerabilities and potential espionage risks.

The decision, announced by the Federal Communications Commission (FCC), adds consumer-grade routers manufactured outside the US to its list of restricted equipment, placing them alongside foreign-made drones, which were banned last year.

The move does not affect routers already in use but applies to all new device models entering the market. Any router built overseas will now require explicit approval before it can be imported, marketed or sold in the US.

Regulators say the decision reflects growing evidence that internet routers, which sit at the heart of home and business networks — have become a key entry point for cyberattacks.

“Malicious actors have exploited security gaps in foreign-made routers to attack American households, disrupt networks, enable espionage, and facilitate intellectual property theft,” the FCC said.

The agency pointed to a series of cyber incidents between 2024 and 2025, known as Volt, Flax and Salt Typhoon, in which compromised networking equipment was allegedly used to target US infrastructure. Investigations by US authorities have linked the attacks to actors associated with the Chinese government.

Under the new framework, manufacturers producing routers outside the US must apply for conditional approval. This process will require companies to disclose foreign ownership or influence and outline plans to shift production to the United States.

Exemptions may be granted in limited cases if equipment is cleared by national security bodies such as the Department of Defense or Department of Homeland Security, although no specific devices have yet been approved.

The ban applies regardless of where a product is designed, meaning even US-based brands that manufacture abroad will be affected.

The decision has significant implications for the global electronics supply chain. The vast majority of consumer routers are currently produced outside the US, particularly in China and Taiwan.

Popular brands such as TP-Link, a major global supplier, have already faced scrutiny amid concerns over cybersecurity vulnerabilities. Even US companies like Netgear, which manufacture overseas, may need to adapt their supply chains to comply with the new rules.

One notable exception is the WiFi router produced by SpaceX’s Starlink service, which the company says is manufactured in Texas.

The move is the latest step in a broader effort by the US to reduce reliance on foreign-made technology deemed critical to national infrastructure. It reflects a growing emphasis on supply chain security and domestic production, particularly in sectors linked to communications, defence and data.

Analysts say the policy could accelerate a wider decoupling in global technology markets, as governments increasingly prioritise security over cost efficiency.

For consumers and businesses, the immediate impact may be limited, but over time the shift could reshape pricing, availability and innovation in networking equipment.

As cybersecurity threats continue to evolve, the US government’s message is clear: devices at the core of digital infrastructure are now considered strategic assets, and their origin matters.

Read more:
US bans new foreign-made routers over national security fears

Teenage darts sensation Luke Littler has applied to trademark his own face in a landmark move aimed at protecting his image from AI-generated fakes and unauthorised commercial use.

The 19-year-old, already a two-time World Darts Championship winner, has submitted an application to the UK Intellectual Property Office as concerns grow over the rapid rise of deepfakes and AI-generated content exploiting public figures.

Littler’s likeness is already widely used across commercial products, from branded dartboards and video games to food items, reflecting his meteoric rise as one of the most marketable names in British sport. He has previously secured trademark protection for his nickname “The Nuke” in the United States, underlining the increasing value of his personal brand.

The latest move signals a growing trend among high-profile athletes and celebrities seeking to protect their identity in an era where AI tools can replicate faces and voices with alarming accuracy.

Graeme Murray, a trademark attorney at Marks & Clerk, said such applications are becoming more common as public figures attempt to safeguard their image. He noted that AI-generated content poses a “genuine threat” to the commercial value and goodwill associated with well-known individuals.

“The objective is to create exclusivity around a recognisable appearance that consumers associate with one individual,” he explained. “This prevents third parties from exploiting that identity without consent, particularly in commercial settings.”

The legal landscape, however, remains uncertain. Unlike some jurisdictions, the UK does not recognise a formal “right of personality”, meaning individuals have limited protection over the commercial use of their likeness outside existing intellectual property frameworks.

Iain Connor, intellectual property partner at Michelmores, warned that trademarking a face is not a comprehensive solution. “Even if successful, trade mark protection is limited to specific categories of goods and services,” he said. “It is not a silver bullet against deepfakes.”

He added that previous attempts to protect identity through trademarks have produced mixed results, citing successful and unsuccessful cases involving public figures. The challenge lies in proving that a face or likeness functions as a distinctive commercial identifier.

The move comes as policymakers and legal experts increasingly debate how to regulate AI-generated content. The UK government has already acknowledged potential gaps in existing copyright and IP frameworks, with discussions underway about introducing new “personality rights” to better protect individuals from digital replication.

Littler’s application therefore represents not only a commercial strategy but also a test case for how far current intellectual property law can stretch in the age of generative AI.

Away from the courtroom, Littler continues to dominate on the oche. Fresh from a dramatic comeback victory over Gerwyn Price in Dublin, he admitted he is still adapting to the pressures of fame and fan scrutiny.

But as his profile continues to grow, so too does the need to protect it, not just from rivals on the darts circuit, but from the increasingly sophisticated capabilities of artificial intelligence.

Read more:
Luke Littler moves to trademark his face in bid to combat AI fakes

High street bank lending to UK businesses climbed to £17.5 billion in 2025, marking a second consecutive year of growth and underlining the increasingly central role of small firms in driving the country’s AI-led economic transformation.

New figures from UK Finance show gross lending rose from £16.1 billion in 2024, with momentum building steadily throughout the year. In the final quarter alone, £4.6 billion was lent, extending the run of year-on-year growth to eight successive quarters.

The increase was driven overwhelmingly by smaller businesses, particularly those with annual turnover of up to £2 million. Lending to this segment rose by more than 25 per cent compared with the previous year, reflecting both stronger demand and improving approval rates.

By contrast, medium-sized firms recorded more modest growth of 4 per cent, suggesting that while confidence is returning across the SME landscape, the smallest businesses are currently the most active in seeking finance.

The data points to a broad-based recovery in business lending across the UK, with activity spread evenly across regions. It also signals a shift in how companies are financing growth, with new loan approvals outpacing overdraft usage, reversing a trend seen in 2024.

However, despite the uptick in lending, utilisation of overdraft facilities remains below pre-pandemic levels, indicating that many businesses are still maintaining financial buffers amid ongoing economic uncertainty, rising costs and geopolitical volatility.

David Raw, Managing Director of Commercial Finance at UK Finance, said the figures highlighted the resilience and importance of the SME sector.

“SMEs are a vitally important part of the UK economy and the banking sector is proud to support them,” he said. “It was good to see gross lending increasing for another consecutive year of growth in 2025, driven by stronger demand from the smallest businesses and support from high street lenders.”

Yet industry experts have warned that while the headline figures are encouraging, they fall well short of what is required to support the next phase of economic growth, particularly as businesses race to adopt artificial intelligence and digital technologies.

Raj Abrol, chief executive of data intelligence firm Galytix, said that structural barriers within traditional banking models continue to limit access to capital for scale-ups and high-growth firms.

“It’s encouraging to see SME lending on the rise, but these figures are a drop in the ocean compared to the actual amounts needed to reboot the global economy,” he said.

“Scale-up companies rely on support from banks to invest in new technology, expand into new markets and hire talent, yet far too many struggle to secure the support they need due to outdated operating models and risk profiling.”

Abrol argued that artificial intelligence itself could play a role in addressing these inefficiencies, particularly in streamlining lending processes and improving credit assessments.

“AI agents can change all this, they do not get tired, do not miss details and do not forget what they learned last quarter,” he said. “They can rapidly prepare loan applications for approval and improve access to finance for SMEs.”

The link between access to capital and AI adoption is becoming increasingly clear, with smaller firms under pressure to invest in automation, data analytics and digital infrastructure to remain competitive.

Kenny MacAulay, chief executive of accounting platform Acting Office, said that without sufficient funding, many SMEs risk being left behind in the technological shift.

“Without access to finance, SMEs will fall drastically behind in the race for AI adoption, which in turn will hit the economy hard,” he said.

“It’s great to see an uptick in lending at a time when so many organisations are at a crossroads with tech investment, but these numbers don’t even begin to cover what is needed for long-term change.”

He added that closer collaboration between government and lenders would be essential to scale up funding and build an economy capable of supporting widespread AI integration.

The latest figures come at a pivotal moment for the UK economy, with policymakers increasingly focused on productivity, innovation and growth. SMEs, often described as the backbone of the economy, are now emerging as a critical engine of that transformation.

However, with borrowing costs still elevated and economic uncertainty lingering, the challenge for both lenders and government will be ensuring that access to finance keeps pace with ambition.

If it does not, the risk is that the UK’s AI-driven growth story could be constrained not by a lack of innovation, but by a lack of capital.

Read more:
SME lending rises to £17.5bn as small businesses drive AI-led growth across the UK

Legal technology company Harvey has signed Suits actor Gabriel Macht as a brand ambassador in an unusual move for the B2B sector, as competition intensifies in the fast-growing legal AI market.

Macht, best known for playing high-powered corporate lawyer Harvey Specter in the hit TV series, will partner with the company whose name was directly inspired by his on-screen character. The role marks a rare crossover between entertainment and enterprise software, where celebrity endorsements remain relatively uncommon.

While consumer technology brands have long leveraged star power, from Apple’s collaborations with musicians to Logitech’s campaigns featuring Hollywood actors, such partnerships are far less typical in enterprise-focused industries such as legal technology. However, the move signals a shift as AI firms seek broader brand recognition in an increasingly crowded market.

Macht said his decision to work with Harvey was rooted in the company’s trajectory and its approach to responsible AI deployment.

“I’m partnering with Harvey because I care about where this company goes,” he said. “I want to support a responsible approach that keeps public interest in view. Harvey’s momentum over the last three-plus years has made it a leading legal AI platform, helping teams change the way they work with AI, faster and with more clarity.”

Founded to bring generative AI into legal workflows, Harvey has rapidly gained traction among law firms and corporate legal departments looking to automate research, contract analysis and document drafting. Its growth reflects a broader shift across the legal profession, where firms are under pressure to improve efficiency while maintaining accuracy and compliance.

The partnership also coincides with the launch of Harvey’s new Instagram presence, @askharvey, as the company looks to build a more visible and accessible brand identity beyond traditional enterprise sales channels.

Winston Weinberg, co-founder and chief executive of Harvey, said Macht’s association with the legal profession made him a natural fit for the company’s next phase of growth.

“Gabriel’s legendary performance as a lawyer continues to inspire people to pursue law,” he said. “There’s no better spokesperson to support Harvey’s global brand growth and the launch of our Instagram account.”

The announcement follows a growing trend of brand-building across the legal AI sector. Earlier this month, rival platform Legora entered into a sponsorship agreement with Swedish golfer Ludvig Åberg, placing its branding in a sporting context more typically associated with consumer-facing companies.

Such moves highlight how even highly specialised software firms are increasingly adopting marketing strategies borrowed from consumer industries, as they compete not just on product capability but on visibility, trust and cultural relevance.

For Harvey, the alignment with a character synonymous with confidence, precision and legal excellence is likely to resonate with a profession navigating rapid technological change. Whether that translates into tangible commercial advantage remains to be seen, but the signal is clear: legal tech is no longer content to operate quietly in the background.

As artificial intelligence continues to redefine how legal work is delivered, firms like Harvey are not only racing to build the most capable platforms, but also the most recognisable brands.

Read more:
Legal AI firm Harvey signs Suits star Gabriel Macht in rare B2B brand deal

The UK government has stepped back from one of its most controversial proposals on artificial intelligence and copyright, signalling a decisive shift towards market-led licensing and greater transparency rather than sweeping legal reform.

In its long-awaited Report on Copyright and Artificial Intelligence, published in March 2026, ministers confirm they will no longer pursue a broad copyright exception for AI training with an opt-out mechanism — a policy that had triggered fierce opposition from across the UK’s creative industries.

Instead, the government is opting for a more cautious, evidence-led approach, prioritising transparency obligations and allowing a nascent but rapidly expanding licensing market to develop. The move marks a significant recalibration of policy at a time when the UK is seeking to position itself as both an AI superpower and a global creative hub.

At the heart of the report is a clear admission: the government’s preferred option, allowing AI developers to use copyrighted material unless rightsholders explicitly opted out, failed to win support.

The consultation attracted more than 11,500 responses, with the overwhelming majority of creators, publishers and rights organisations rejecting the proposal outright.

Ministers now concede that a broad exception “with opt-out is no longer the government’s preferred way forward”, citing strong industry opposition, lack of consensus, and insufficient evidence on economic impact.

This represents a notable victory for the UK’s creative sectors, from publishing and music to film and photography, which argued that such an exception would effectively legalise uncompensated use of their work by generative AI systems.

The report lays bare the fundamental policy dilemma: how to balance AI-driven economic growth with the protection of intellectual property.

On one side sit AI developers, who require vast datasets, often including copyrighted material, to train large language models and generative systems. On the other are creators whose works underpin those systems but risk being displaced by them.

The government acknowledges that modern AI models are typically trained on “billions of copyright works”, raising complex questions about fairness, consent and competition.

Yet it also highlights uncertainty around the economic benefits of reform, noting limited evidence that loosening copyright rules would materially increase AI investment in the UK.

In effect, ministers are choosing to pause rather than gamble.

Rather than legislating, the government is placing its bets on licensing, a market-based mechanism already beginning to take shape.

A growing number of deals between AI firms and content owners, particularly in publishing, music and image libraries, suggests a commercial model is emerging. However, the report acknowledges this market is still “new and evolving” and lacks transparency.

Crucially, ministers have ruled out direct intervention for now:

“We propose not to intervene in the licensing market at this stage… and will keep market-led approaches under review.”

This position aligns closely with industry sentiment across both creative and technology sectors, which broadly favour voluntary, negotiated agreements over statutory schemes.

However, it also raises important questions, particularly for SMEs and individual creators, about bargaining power and equitable remuneration.

Among those welcoming the shift is Tom West, CEO of Publishers’ Licensing Services (PLS), who sees licensing as both practical and scalable.

West said: “We welcome that the government has listened to the strong response it received from across the UK’s creative industries to its consultation and has stepped back from its preferred option of a copyright exception with an opt out and is to review the transparency of AI inputs, which would further boost licensing.

Whilst we await further clarity from the government on the long-term direction of its copyright policy, PLS will continue to serve our publishers and work with our partners on market-based, industry-backed AI licensing solutions.

This approach is already being put into practice. At the London Book Fair last week, PLS launched the first stage of a new collective licensing solution designed specifically to support the use of published content in AI. It was met with strong interest and positive feedback from publishers and industry partners, with publishers already beginning to sign up. The solution offers a practical, scalable way for AI developers to access high-quality content while ensuring creators are paid and retain control over how their work is used.

The case has not been made for the introduction of a new copyright exception. There is no market failure and a dynamic licensing market for the use of content in AI has developed and continues to grow. Any copyright exception for generative AI would jeopardise these licensing solutions, removing the ability of large and small rightsholders to receive payment for the use of their works in AI and reducing control over their content.

PLS welcomes the government’s engagement on this critical issue. We share a commitment to a mutually beneficial outcome and invite the government to work closely with us to help further develop and promote licensing options that support rightsholders of all sizes and AI developers seeking high-quality, trusted content.”

If licensing is the economic mechanism, transparency is the regulatory lever.

More than 90% of consultation respondents supported requirements for AI developers to disclose the sources of training data.

The government agrees, in principle, but stops short of immediate regulation. Instead, it proposes:
• developing industry-led best practice
• monitoring international frameworks (notably the EU AI Act)
• considering future legislation if needed

Transparency is seen as essential to enable enforcement, licensing and trust, particularly given that creators often have no visibility over whether their work has been used.

For UK businesses, particularly SMEs, the implications are nuanced.

For creators and publishers
• greater protection in the short term
• stronger negotiating position in licensing deals
• ongoing challenges around enforcement and visibility

For AI startups and developers
• continued legal uncertainty
• potential cost barriers to accessing training data
• reliance on licensed or overseas-trained models

For the wider economy
• slower regulatory clarity
• reduced risk of over-regulation
• continued dependence on global AI ecosystems

The report explicitly notes that SMEs on both sides, creators and developers, face disproportionate challenges under the current system.

Perhaps the most striking aspect of the report is its tone: cautious, iterative, and deliberately non-committal.

The government repeatedly emphasises the need for more evidence, more international alignment, and more market development before taking decisive legislative action.

With ongoing litigation in the US, new rules emerging in the EU, and rapid advances in generative AI, the UK risks being pulled in multiple directions, economically, legally and politically.

This is not a resolution, it is a holding position.

By stepping back from sweeping reform, the government has bought time. But it has also shifted responsibility onto the market to prove that licensing can work at scale, fairly and efficiently.

If it can, the UK may yet carve out a balanced model that supports both innovation and creativity.

If it cannot, the debate over copyright and AI will return, sharper, louder, and far harder to resolve.

Read more:
UK government backs away from AI copyright overhaul as licensing emerges as the battleground

Trustpilot has emerged as an early beneficiary of the shift towards artificial intelligence-led search, reporting a sharp rise in profits and a surge in its share price after a year of strong growth driven by increased exposure through large language models.

Shares in the consumer review platform jumped as much as 28 per cent following results that beat market expectations, as investors responded positively to signs that the business is successfully adapting to the changing dynamics of online discovery. The company posted pre-tax profits of $14.1 million for the year to December, up significantly from $5.2 million the previous year, underpinned by stronger customer retention and a shift towards higher-value contracts.

Revenue rose 24 per cent year-on-year, with growth recorded across the UK, Europe and the United States. Trustpilot also reported a 16 per cent increase in average annual contract value, reflecting its success in moving upmarket and monetising its platform more effectively.

Central to that performance has been the company’s growing visibility within AI-powered search environments. Trustpilot said click-throughs from AI-driven platforms increased more than fifteenfold over the past year, highlighting how rapidly consumer behaviour is shifting away from traditional search engines towards conversational interfaces powered by large language models.

The company has actively opened its data to these platforms, allowing its reviews to be surfaced within AI-generated answers. According to Promptwatch data, Trustpilot ranked as the fifth most cited domain globally on ChatGPT in January, a position that has significantly enhanced its reach and relevance.

Chief executive Adrian Blair described artificial intelligence as a “major tailwind” for the business, noting that visibility within AI search has become a key selling point when engaging with clients. As businesses increasingly focus on how they appear within AI-generated responses, Trustpilot’s repository of verified consumer feedback has become a valuable asset in the emerging search ecosystem.

Analysts suggested the results offer an early indication that the transition from traditional search to AI-led discovery could create new winners, particularly for platforms built around user-generated content. Investec analysts noted that Trustpilot’s performance demonstrates how this shift could benefit businesses whose data is highly relevant to AI-driven queries.

Alongside its earnings growth, Trustpilot announced a £30 million share buyback programme, including £7.5 million allocated to its employee benefit trust, signalling confidence in its financial position and long-term prospects. The company also upgraded its medium-term profitability targets, forecasting that its adjusted EBITDA margin will rise from 15.6 per cent in 2025 to 25 per cent by 2028 and 30 per cent by 2030.

The strong results mark a rebound after a turbulent period for the company’s share price. In December, Trustpilot faced scrutiny following claims by short-seller Grizzly Research alleging questionable practices in its dealings with non-paying customers. The company strongly denied the allegations and issued a detailed rebuttal, helping to stabilise investor sentiment after an initial sell-off.

The stock was also caught in a broader downturn affecting software companies earlier this year, but the latest results suggest Trustpilot may be structurally better positioned than many peers in an AI-driven market.

Blair emphasised that the company’s core proposition remains fundamentally distinct from other technology businesses. While AI can aggregate and present information, he argued, it cannot replicate the real-world customer experiences that underpin Trustpilot’s platform.

As artificial intelligence continues to reshape how consumers search, discover and evaluate brands, Trustpilot’s ability to embed itself within that ecosystem appears to be driving both immediate performance gains and longer-term strategic value.

Read more:
Trustpilot profits soar as AI-driven traffic fuels sharp share price rally

The Chancellor, Rachel Reeves, is set to unveil a sweeping £2 billion investment programme aimed at positioning the UK as the fastest adopter of artificial intelligence across the G7, in a major bid to reignite economic growth and strengthen Britain’s global competitiveness.

In a keynote speech in London, Reeves will outline a strategy centred on accelerating AI adoption across both the private and public sectors, with the government taking a more interventionist role in shaping how emerging technologies are deployed. The move signals a shift towards a more active industrial policy, with AI at its core.

The funding package will be directed not only at advancing AI capabilities but also at building the infrastructure required to support large-scale deployment. Central to this is expanded access to high-performance computing and the development of a national quantum computing programme, designed to unlock faster processing power and enable breakthroughs in areas such as healthcare diagnostics, energy optimisation and secure communications.

Reeves is expected to argue that the UK must play a leading role in determining how AI evolves globally, rather than adopting frameworks set by other nations. Framing the initiative as both an economic and strategic imperative, she is set to warn against ceding influence in a technology that is rapidly reshaping industries and societies.

The strategy will focus heavily on accelerating real-world adoption. Ministers plan to work closely with businesses, academic institutions and investors to embed AI tools across key sectors, from financial services and manufacturing to healthcare and local government. The NHS, in particular, is expected to be a major beneficiary, with AI applications aimed at improving efficiency, diagnostics and patient outcomes.

A significant component of the plan will also be regional growth. The government is expected to highlight initiatives such as the Oxford-Cambridge innovation corridor as focal points for AI development, talent cultivation and start-up activity, with the aim of spreading economic benefits beyond London and the South East.

The announcement comes at a time when the UK economy is facing subdued growth and rising global uncertainty, with policymakers increasingly looking to technology-driven productivity gains as a route to long-term expansion. By positioning AI as a foundational economic driver, Reeves is seeking to create what she has previously described as a “strategic and active state” approach to growth.

However, industry experts have cautioned that rapid adoption must be matched with robust governance and infrastructure. Stuart Harvey, chief executive of data specialist Datactics, warned that organisations are already beginning to rely on AI for decision-making, raising concerns about transparency and accountability.

He noted that without strong data foundations and auditability, AI-driven decisions could become opaque and difficult to challenge, particularly in high-stakes environments such as public policy or corporate strategy. The risk, he suggested, is that poorly governed adoption could lead to significant economic and societal consequences.

Similarly, Sachin Agrawal, managing director of Zoho UK, welcomed the ambition but stressed the importance of targeted investment. He argued that success would depend on strengthening digital infrastructure, developing regional talent pipelines and ensuring responsible data management.

Agrawal also highlighted the critical role of regulation, suggesting that effective oversight should not be seen as a barrier to innovation but as a means of building trust. Without clear standards and accountability, he warned, public confidence in AI could be undermined by high-profile failures.

The government is expected to acknowledge these challenges, with plans to support workforce reskilling and ensure that regulatory frameworks evolve alongside technological advancements. As AI becomes more deeply embedded in business operations and public services, the pressure on policymakers to balance innovation with oversight is set to intensify.

Reeves’ announcement reflects a broader global race to harness artificial intelligence as a driver of productivity and economic growth. With the US and China investing heavily in AI infrastructure and development, the UK’s strategy is designed to ensure it remains competitive on the world stage.

Whether the £2 billion investment will be sufficient to achieve that ambition remains an open question. However, the signal from government is clear: AI is no longer a peripheral policy area, but a central pillar of the UK’s economic future.

Read more:
Rachel Reeves unveils £2bn AI push to make UK fastest adopter in G7

Chinese e-commerce giant JD.com has made a decisive move into the UK market with the launch of its Joybuy platform, setting up a direct challenge to Amazon by promising same-day delivery without the traditional trade-off between speed and price.

The new platform marks JD.com’s most significant expansion into Britain to date, following years of speculation about its ambitions in the market. Joybuy, which had previously been tested through a London pilot, is now rolling out more widely, offering British consumers access to a broad product range spanning electronics, groceries, gaming, household goods and everyday essentials.

The retailer is positioning Joybuy as a full-spectrum marketplace, stocking global brands such as Apple, Samsung and Sony alongside consumer staples including Heinz, Cadbury and Coca-Cola. The proposition is clear: convenience at scale, backed by logistics infrastructure designed to rival, and potentially outpace, incumbents.

At the core of the launch is JD.com’s “Double 11” delivery promise. Orders placed before 11am will be delivered by 11pm the same day, with free delivery available on orders over £29. The company said the service will initially cover more than 17 million consumers across key urban centres including Birmingham, Leicester and Nottingham, signalling a deliberate focus on high-density, high-demand regions.

This logistics-led strategy reflects JD.com’s long-established operating model in China, where it has built one of the most vertically integrated fulfilment networks in global e-commerce. Rather than relying heavily on third-party couriers, the group controls much of its supply chain, from warehousing to last-mile delivery, enabling tighter control over speed, cost and customer experience.

In the UK, that model is being replicated through JoyExpress, the company’s delivery arm, which is supported by a growing European infrastructure footprint. JD.com already operates more than 60 warehouses and depots across Europe, including key UK sites in Milton Keynes and Luton, providing the backbone for its same-day ambitions.

A spokesperson for Joybuy said the company aims to “change the way people shop online” by removing the longstanding compromise between affordability and delivery speed. “British shoppers have long had to settle for a trade-off between price and speed,” they said. “We’re here to change that.”

The expansion comes at a time when JD.com is seeking growth outside its domestic Chinese market, where consumer demand has softened and competition has intensified. The company, which has a market capitalisation of more than $40 billion, has been actively exploring international opportunities as part of a broader diversification strategy.

Its interest in the UK is not new. The group previously attempted to acquire Argos from Sainsbury’s and held discussions around a potential deal with Currys, although neither transaction materialised. The Joybuy launch represents a shift from acquisition-led expansion to organic market entry, allowing JD.com to build its presence on its own terms.

However, analysts caution that replicating its Chinese logistics model in Europe will not be straightforward. The UK’s fragmented retail landscape, regulatory environment and established competition present significant barriers to scaling quickly. Amazon, in particular, retains a dominant position, underpinned by its Prime ecosystem, extensive fulfilment network and deep customer loyalty.

Even so, JD.com’s entry introduces a new competitive dynamic into the UK e-commerce market. Its willingness to invest heavily in infrastructure and absorb delivery costs could place pressure on incumbents, particularly if consumers respond positively to faster delivery without additional fees.

The move also reflects a broader shift in online retail, where speed is increasingly becoming a key differentiator. As consumer expectations evolve, same-day delivery is moving from a premium offering to a baseline expectation in major urban markets.

JD.com’s chairman, Liu Qiangdong, has previously acknowledged that the company has faced a challenging period in recent years, describing the past five years as the least productive of his entrepreneurial career. The UK launch of Joybuy suggests a renewed push for growth, and a belief that international markets can provide the next phase of expansion.

For British consumers, the arrival of Joybuy could signal the start of a new era in e-commerce competition — one where delivery speed, pricing and platform experience are being redefined simultaneously.

Read more:
JD.com launches Joybuy in UK with same-day delivery challenge to Amazon

The UK government has unveiled a new online tool designed to help households and businesses track the rollout of gigabit-capable broadband across the country, offering greater transparency over when faster connectivity will reach local communities.

The new address checker allows users to enter their postcode and see whether their property is scheduled to receive an upgrade through the government’s Project Gigabit programme or through separate commercial full-fibre deployments. Officials say the tool is intended to provide rural communities and businesses with clearer visibility of broadband infrastructure plans, particularly in areas where connectivity improvements have historically been slow.

The launch forms part of the government’s wider effort to accelerate the delivery of high-speed broadband across the UK, with particular emphasis on rural and hard-to-reach regions that have traditionally struggled with poor digital infrastructure.

According to the Department for Science, Innovation and Technology, more than 750 homes and businesses are now gaining access to gigabit-capable broadband every day through Project Gigabit. The programme is designed to deliver full-fibre connectivity to areas that are unlikely to be served by commercial investment alone.

Officials estimate that more than one million additional premises will benefit from live government contracts currently being rolled out across rural England and Wales. These include major infrastructure agreements with broadband providers aimed at expanding fibre networks into remote towns, villages and agricultural communities.

The government argues that improving digital connectivity is critical to supporting economic development outside major cities. Faster broadband access is expected to enable remote working, improve access to digital public services and strengthen sectors such as agriculture, tourism and rural small businesses.

However, campaigners warn that improving infrastructure alone will not eliminate the UK’s digital divide.

Elizabeth Anderson, chief executive of the Digital Poverty Alliance, said that while expanding gigabit broadband coverage is an important milestone, affordability remains a major barrier for millions of people.

“The continued rollout of gigabit-capable broadband and improved mobile coverage in rural communities is a welcome step towards closing long-standing connectivity gaps across the UK,” she said.

“However, infrastructure alone will not solve digital poverty. Around 19 million people in the UK experience some form of digital exclusion, and government figures show that around 1.6 million people are still living entirely offline.”

She added that the cost of broadband services and suitable devices continues to prevent many households from accessing digital services.

“We estimate around two million people lack connectivity because of affordability, and gigabit broadband is frequently out of reach due to higher costs,” Anderson said.

“While faster networks are important, they only make a difference if people can afford to use them. Connectivity must be not only available, but affordable and accessible for everyone.”

Alongside fibre expansion, the government is also investing in improved mobile connectivity through the Shared Rural Network, a joint initiative between government and the UK’s major mobile network operators.

The programme aims to extend 4G coverage into rural “not-spots”, areas where reliable mobile signals have historically been unavailable. Recent upgrades have already expanded coverage significantly across parts of the UK countryside.

Industry experts say demand for postcode-level fibre availability tools has grown significantly as rollout accelerates across the UK. Tomas Novosad, founder of UK fibre availability platform Full Fibre Checker, said consumers increasingly want clearer visibility into when full fibre will reach their homes.

“We’re seeing more people actively checking availability and rollout timelines rather than waiting for providers to announce upgrades. As Project Gigabit expands, postcode and address-level tools play an important role in helping households understand whether improvements are coming soon or if they should consider alternative providers,” Novosad said.

Jennifer Holmes, chief executive of the London Internet Exchange (LINX), said the continued expansion of gigabit broadband and mobile coverage represents a key step in strengthening the UK’s digital infrastructure.

“As demand for online services continues to grow, the networks that underpin the internet must be resilient, efficient and capable of supporting increasing volumes of data,” she said.

“Strong infrastructure is essential not only for everyday connectivity, but also for supporting innovation, economic growth and the UK’s wider digital ambitions.”

Holmes added that modern digital networks now underpin almost every part of the economy, from cloud computing and artificial intelligence to e-commerce and public services.

“Investment in faster and more reliable connectivity will help ensure that businesses, public services and communities can fully participate in an increasingly digital economy,” she said.

The new postcode tool is intended to give consumers and businesses clearer information about when gigabit broadband will reach their homes or workplaces, particularly in areas where rollout timelines have previously been unclear.

By providing greater transparency over rollout plans, ministers hope the tool will help local communities better plan for the future and encourage businesses to invest in rural areas with improved connectivity.

Project Gigabit remains one of the UK government’s flagship infrastructure initiatives, aimed at ensuring that the vast majority of UK premises have access to gigabit-capable broadband by the end of the decade.

But as rollout accelerates, policymakers and campaigners alike warn that bridging the digital divide will require more than infrastructure alone. Ensuring that connectivity is affordable, accessible and supported by digital skills programmes will be crucial if the benefits of the UK’s digital transformation are to be shared across every community.

Read more:
Government launches postcode checker to track gigabit broadband rollout across UK

The UK government has announced a new package of initiatives designed to boost female participation in the technology sector, including a £4 million programme to support hundreds of women into tech jobs and inspire thousands of schoolgirls to pursue digital careers.

The measures form part of a broader strategy aimed at addressing the persistent gender imbalance across the technology industry, where women remain significantly underrepresented despite the sector’s rapid growth and importance to the UK economy.

Ministers say the initiatives are intended to help women enter, remain in and return to technology roles, while also encouraging more girls to consider technology careers earlier in their education.

At the centre of the announcement is the new TechFirst Women’s Programme, backed by £4 million of government funding. The scheme aims to create at least 300 paid placements in technology roles across the UK.

The programme will work with businesses, including small and medium-sized enterprises, to identify opportunities for women to gain experience in fields such as software development, digital engineering, data science and artificial intelligence.

Participants will receive career coaching, interview preparation and technical support to help them secure roles and progress within the sector.

The government hopes the initiative will not only help individuals advance their careers but also support companies looking to adopt new technologies, particularly artificial intelligence, by giving them access to skilled workers.

Ministers say the programme addresses a significant economic issue. Research suggests the UK loses between £2 billion and £3.5 billion each year due to women leaving the technology sector.

Alongside the placement programme, the government is launching a pilot returnship scheme aimed at helping experienced developers re-enter the workforce after career breaks.

The initiative will initially operate within government departments including the Home Office and the Ministry of Justice.

The programme will target skilled software developers who have been out of work for at least 18 months, a group that includes many women who have taken time away from employment to care for children or family members.

Participants will be supported in returning to senior technology roles within the public sector, with the aim of tackling what many professionals refer to as the “CV gap” barrier that can prevent experienced workers from returning to their careers.

Officials say the scheme could later be expanded across other departments or into the private sector if successful.

The government is also attempting to address the gender gap earlier in the talent pipeline through a new national technology competition aimed at schoolgirls.

Later this year, thousands of girls aged 12 and 13 will be invited to take part in the TechFirst Girls Competition, a nationwide initiative designed to introduce students to coding, artificial intelligence and digital problem-solving.

The competition will challenge participants to develop creative solutions to real-world problems using technology, while also offering insight into how digital skills translate into future careers.

Technology company IBM will partner with the Department for Science, Innovation and Technology to deliver the programme.

The initiative builds on previous efforts such as the CyberFirst Girls Competition, which has already involved more than 10,000 students across the UK.

As part of the wider policy drive, the government’s Women in Tech Taskforce has launched a call for evidence to better understand the challenges women face when entering or progressing in the technology industry.

The consultation will gather insights from workers, businesses and industry organisations about issues ranging from recruitment practices to workplace culture and career progression.

The taskforce will also examine how emerging technologies such as artificial intelligence may be reinforcing existing inequalities.

Research has already shown that AI systems trained on historical employment data can replicate past biases. For example, some recruitment algorithms have been found to favour male candidates over female applicants when evaluating job applications.

Officials say gathering real-world experiences will help shape future policy designed to make the technology sector more inclusive.

Announcing the programme, Technology Secretary Liz Kendall said women continue to face significant barriers within the industry.

“I am very aware of the reality women face in tech: women aren’t being given a fair shot, whether that’s getting into the sector, staying in it, or returning after time away,” she said.

“If we don’t address these biases and barriers now, we’ll still be having this conversation in ten years’ time.”

Kendall said the government wanted to ensure women were not only entering the industry but also shaping its future.

“These aren’t warm words, they’re real jobs, real placements and real routes back in through a door that has been too hard to open for too long.”

Industry figures welcomed the announcement but emphasised that long-term structural change would be needed to close the gender gap in technology.

Anna Brailsford, chief executive of Code First Girls and a member of the Women in Tech Taskforce, said improving access to training and career opportunities could transform lives.

“Many women who have moved into tech started their journeys unsure if they belonged in the industry and are now thriving in high-impact roles,” she said.

“The UK’s ambition to lead in technology will only be realised if more women can see a clear and supported pathway into the sector from non-tech backgrounds.”

The government hopes the new measures will help strengthen the UK’s technology workforce at a time when demand for digital skills continues to grow rapidly across industries ranging from finance and healthcare to manufacturing and artificial intelligence development.

Read more:
Government launches new programme to help more women and girls enter the UK tech sector

Artificial intelligence company Anthropic has filed an unprecedented lawsuit against the United States government after being formally labelled a “supply chain risk”, escalating a bitter dispute over the military use of advanced AI technology.

The legal action, filed in a federal court in California, challenges a directive issued by the administration of Donald Trump that effectively barred US government agencies from using Anthropic’s AI systems. The company argues the move was politically motivated retaliation after it refused to remove restrictions on how its technology could be deployed by the US military.

Anthropic’s lawsuit claims the decision was “unprecedented and unlawful” and violated constitutional protections around free speech and due process.

“The Constitution does not allow the government to wield its enormous power to punish a company for its protected speech,” the firm said in its complaint. “No federal statute authorises the actions taken here.”

The conflict stems from a disagreement between Anthropic’s chief executive Dario Amodei and US defence officials, including Pete Hegseth, over how the company’s artificial intelligence tools could be used by the Pentagon.

Anthropic has long maintained strict contractual limits on the deployment of its technology, including bans on using its AI models for “lethal autonomous warfare” and for mass domestic surveillance of American citizens.

According to the lawsuit, defence officials demanded that the company remove these restrictions from its government contracts. Anthropic refused, arguing that such safeguards were essential to ensure responsible use of powerful AI systems.

The company said negotiations with the Department of Defense were initially progressing and that both sides had been working toward revised language that would allow continued cooperation while preserving ethical limits.

However, those talks reportedly collapsed after the White House intervened.

Following the breakdown in negotiations, the Pentagon designated Anthropic as a “supply chain risk” — a classification normally applied to companies considered insecure or unreliable partners for government systems.

The designation effectively blocks US government agencies and contractors from using Anthropic’s software tools.

The move was accompanied by public criticism from the Trump administration, with White House officials accusing the company of attempting to dictate military policy.

Liz Huston, a spokesperson for the White House, told reporters that Anthropic was “a radical left, woke company” seeking to impose its own conditions on national defence operations.

“Under the Trump Administration, our military will obey the United States Constitution — not any woke AI company’s terms of service,” Huston said.

Anthropic disputes that characterisation and argues that its restrictions were standard contractual provisions designed to prevent misuse of AI systems.

The legal challenge names a broad list of defendants, including the executive office of President Trump and senior government officials such as Marco Rubio and Howard Lutnick.

The suit also targets 16 federal agencies, including the Departments of Defense, Homeland Security and Energy.

Anthropic claims the directive banning its technology has caused significant reputational and commercial damage.

The company said that both current and prospective commercial contracts were now under threat, potentially jeopardising “hundreds of millions of dollars in the near term”.

It also argued that the decision had created a broader chilling effect across the technology sector by discouraging companies from speaking publicly about the risks associated with advanced AI.

The case has already drawn support from across the technology industry.

Nearly 40 employees from rival companies including Google and OpenAI filed a joint legal brief backing Anthropic’s position, despite the firms being competitors in the rapidly expanding AI sector.

The signatories warned that the deployment of advanced AI systems without safeguards could create serious risks, particularly if used for mass surveillance or autonomous weapons.

“As a group, we are diverse in our politics and philosophies,” the engineers wrote in their submission. “But we are united in the conviction that today’s frontier AI systems present risks when deployed to enable domestic mass surveillance or the operation of autonomous lethal weapons systems without human oversight.”

Anthropic’s flagship AI system, Claude, has become widely used by technology companies and developers for coding, research and enterprise software tasks.

Companies such as Microsoft, Amazon and Meta have confirmed they will continue to use the technology in commercial applications, although not in projects involving US defence agencies.

Anthropic is not seeking financial damages in the case. Instead, it is asking the court to declare the government’s directive unconstitutional and remove the “supply chain risk” designation immediately.

Legal experts believe the dispute could become a landmark case in defining how governments interact with AI developers.

Carl Tobias, a law professor at the University of Richmond, said the case could ultimately reach the US Supreme Court.

“Anthropic may very well win in federal court,” Tobias said. “But this administration is not shy about appealing. It will probably go to the Supreme Court.”

The outcome could have major implications for the fast-growing AI industry, particularly as governments worldwide increasingly rely on private technology firms to supply critical artificial intelligence systems for defence, intelligence and national security operations.

For now, the lawsuit marks a rare moment in which a major technology company is openly challenging government authority over the future deployment of artificial intelligence.

Read more:
Anthropic sues US government after being labelled a ‘supply chain risk’ in AI dispute

A growing number of businesses are paying cybercriminals after ransomware attacks, as hackers deploy artificial intelligence to make their tactics more targeted, sophisticated and damaging.

New research from cybersecurity consultancy S-RM and advisory firm FGS Global shows that 24.3 per cent of companies targeted by ransomware attacks paid the demanded ransom in 2025, marking a sharp increase from 14.4 per cent in 2024.

The figures represent the first significant rise in ransom payments after two years of decline. In 2023, about 16.4 per cent of affected organisations paid, while the peak came in 2022 when 27.6 per cent of victims settled with attackers.

Although the latest numbers remain below that high point, the jump suggests cybercriminals are becoming increasingly successful at pressuring companies into handing over money.

Cybersecurity experts say artificial intelligence is rapidly reshaping how ransomware attacks are planned and executed.

Hackers are now able to use AI tools to scan vast amounts of stolen or publicly available data, allowing them to identify the most sensitive information belonging to a target organisation. By focusing on data that could cause the greatest reputational, financial or operational damage if exposed, attackers are able to increase pressure on victims to pay.

Jamie Smith, head of cybersecurity at S-RM, said criminals were increasingly relying on AI to refine their strategies.

“Attackers are using AI to find the most sensitive information that could cause maximum damage,” he said. “Threats are becoming far more specific and personalised, designed to maximise the victim’s fear and willingness to pay.”

This evolution has made ransomware attacks more difficult for companies to defend against, particularly for organisations with large volumes of sensitive data.

The report also sheds light on the scale of payments being demanded by cybercriminal groups.

According to the study, ransom payments in 2025 ranged from as little as $10,000 to more than $1 million, with the average payment reaching $296,000.

However, cybersecurity specialists warn that the total cost of a ransomware attack often extends far beyond the ransom itself. Businesses frequently face operational disruption, regulatory scrutiny, reputational damage and the expensive process of rebuilding compromised IT systems.

Many organisations also incur costs related to legal advice, customer notifications and forensic investigations after an attack.

The research suggests that industrial and manufacturing companies were particularly likely to pay ransoms during the past year.

This trend appears to be driven by the severe operational disruption ransomware attacks can cause in sectors that rely heavily on continuous production.

Factories, logistics systems and supply chains can grind to a halt if core IT infrastructure becomes inaccessible. In such situations, businesses sometimes view paying a ransom as the quickest way to restore operations and avoid prolonged shutdowns.

One high-profile cyber incident involved Jaguar Land Rover, whose factories around the world were forced to shut down for the entire month of September after its IT systems were compromised.

Major UK retailers were also targeted in 2025, including Marks & Spencer and Co-op. None of the companies has publicly confirmed whether a ransom was paid.

One of the biggest challenges in measuring ransomware activity is that many companies refuse to disclose whether they have paid hackers.

Security specialists say businesses often fear that publicly admitting to ransom payments could make them more attractive targets for future attacks.

Criminal groups may interpret payment as a sign that a company has both the resources and willingness to comply with demands.

As a result, ransomware incidents are often kept confidential, with payments handled through private negotiations involving cybersecurity consultants, insurers and specialist crisis advisers.

While artificial intelligence is helping companies automate operations and improve efficiency, experts warn it is also opening up new vulnerabilities that cybercriminals are eager to exploit.

Jenny Davey, co-head of crisis management at FGS Global, described the technology as a “double-edged sword”.

“While AI can drive efficiency and performance across the business, it can also open up new attack vectors for cybercriminals to exploit,” she said.

The rapid adoption of AI tools across corporate systems means organisations must invest heavily in cybersecurity and staff training to avoid creating new entry points for attackers.

The rise in ransomware payments highlights the growing importance of cyber resilience for businesses across every sector.

Experts say companies must go beyond traditional IT security measures and adopt a broader approach that includes employee awareness, robust data protection practices and detailed incident response plans.

This includes maintaining secure backups, limiting access to sensitive information and regularly testing systems against potential cyber threats.

As ransomware attacks become more sophisticated, and increasingly powered by artificial intelligence, businesses face mounting pressure to strengthen their defences before becoming the next target.

Read more:
More companies paying ransoms as AI-powered cyberattacks intensify

Fifty government-funded mobile mast upgrades have now been activated across Wales as part of the UK’s Shared Rural Network (SRN) programme, marking a significant milestone in efforts to improve digital connectivity in some of the country’s most remote communities.

The newly upgraded masts form part of a wider national rollout designed to expand reliable 4G coverage to rural areas that have historically struggled with weak or inconsistent mobile signals. Across the UK, a total of 119 masts funded through the initiative are now live, helping to extend coverage to towns, villages, national parks and major road routes that previously experienced patchy service.

The latest upgrades have been delivered by enhancing existing infrastructure rather than constructing entirely new sites, allowing communities to benefit from stronger mobile coverage while limiting the environmental and planning challenges associated with building additional towers. As a result, residents, visitors and businesses across rural Wales can now access more reliable connectivity without significant changes to the surrounding landscape.

Communities benefiting from the latest phase of the rollout include Ysbyty Ifan, Pentrefoelas, Capel Celyn, Painscastle, Hay-on-Wye, Llanigon, Tregoyd, Doly-y-Gaer, Clwydyfagwyr, Pontsticill, Torpantau, Llanddewi, Dolau, Llandegley, Crossgates and Abbeycwmhir. The improvements also extend into key tourism areas including Eryri National Park and Bannau Brycheiniog National Park, both of which attract millions of visitors each year.

In addition to strengthening coverage in rural settlements, the upgrades provide full 4G access from all four of the UK’s major mobile network operators, EE, Three UK, Virgin Media O2, and Vodafone, across more than 3,400 kilometres of Welsh roads. For many drivers travelling through rural areas, this means improved navigation, communication and access to emergency services in places where signals were previously unreliable.

The Shared Rural Network was first announced in 2020 as a partnership between the UK government and the country’s mobile operators to close the digital divide between urban centres and rural communities. The programme combines £184 million in public funding with more than £500 million of private sector investment from mobile network providers to expand nationwide coverage.

Since the initiative began, 4G coverage from all four operators has expanded significantly, rising from around 66 per cent of the UK’s landmass to approximately 81 per cent. According to programme operator Mova, the expansion represents an area roughly equivalent to the combined size of Wales and Northern Ireland.

Ben Roome, chief executive of Mova, said the Welsh milestone demonstrates the power of collaboration between government and industry in addressing longstanding connectivity gaps.

“Upgrading 50 EAS masts in Wales shows the strength of a shared, neutral programme,” he said. “Every site benefits every operator, every community and every mobile user. Together they represent practical steps toward fairer, more resilient connectivity across rural Wales.”

Improved connectivity is expected to deliver a range of economic and social benefits, particularly for rural businesses and tourism operators that increasingly rely on mobile access for digital services. Reliable 4G coverage can support online bookings for hospitality businesses, enable farmers and rural enterprises to use cloud-based tools, and allow residents to access services such as banking, healthcare and education platforms more easily.

The milestone has also been welcomed by the UK government. Jo Stevens said that improving mobile coverage is an essential part of supporting economic growth and opportunity across rural communities.

“Access to fast and reliable mobile coverage is increasingly important for residents, businesses and community organisations in rural communities all over Wales,” she said. “Hitting this milestone is an important step in our mission to grow the Welsh economy, supporting businesses to succeed and creating opportunities in every corner of Wales.”

Nationwide, the Shared Rural Network programme has already delivered improved connectivity to an additional 280,000 premises and more than 16,000 kilometres of roads. The upgrades focus largely on so-called Extended Area Service masts, which were originally designed to provide coverage from a single operator but are now being modernised so that customers from all networks can benefit.

Further upgrades are planned as the programme continues over the coming years, with the goal of ensuring that even the UK’s most remote communities can access reliable mobile connectivity. For many parts of rural Wales, the activation of these latest sites represents a meaningful improvement in everyday digital access, helping to ensure that residents and businesses are no longer left behind in an increasingly connected economy.

Read more:
Government-funded mobile mast upgrades reach 50 milestone in Wales

OpenAI has confirmed it is postponing the launch of an “adult mode” for ChatGPT, saying the company will instead prioritise improving the platform’s core capabilities and user experience.

The move marks a shift from earlier plans outlined by Sam Altman, who indicated last year that the artificial intelligence developer would allow certain forms of adult content on its flagship chatbot once robust age-verification systems had been introduced.

However, OpenAI has now said that development resources are being redirected toward upgrades that will benefit a broader share of the chatbot’s rapidly expanding user base.

“We’re pushing out the launch of adult mode so we can focus on work that is a higher priority for more users right now,” the company said. “That includes gains in intelligence, personality improvements, personalisation and making the experience more proactive.”

The company added that it still supported the underlying principle behind the proposed feature, allowing adult users greater freedom in how they interact with AI systems, but acknowledged that implementing it safely would require additional work.

“We still believe in the principle of treating adults like adults,” OpenAI said. “But getting the experience right will take more time.”

The decision comes at a time of intense competition in the artificial intelligence sector. Since announcing plans to loosen restrictions on ChatGPT content in late 2025, Altman has repeatedly warned that OpenAI faces a “code red” challenge from rival AI developers.

Among the most prominent competitors are Google DeepMind and Anthropic, both of which are racing to release more capable generative AI systems.

OpenAI’s focus on performance improvements reflects the escalating pressure to maintain leadership in the AI market, where advances in reasoning capability, conversational tone and personalisation are increasingly seen as key differentiators.

The company says ChatGPT now has more than 900 million users worldwide, making it one of the fastest-growing digital platforms in history. Maintaining reliability, safety and usefulness at such scale has become a central priority.

Although the launch of adult mode has been delayed, OpenAI is continuing to develop age-verification and age-prediction systems designed to ensure younger users are protected from inappropriate content.

The technology analyses usage patterns and behavioural signals to estimate whether a user may be under the age of 18. If the system determines that a user is likely to be a minor, stricter safety filters are automatically applied.

These additional safeguards limit exposure to graphic violence, explicit content and sexual role-play scenarios.

The work is also partly driven by regulatory pressures in several countries. In the UK, for example, the Online Safety Act requires platforms hosting potentially harmful or adult material to ensure that under-18s cannot access such content without effective age verification measures.

As a result, any future “adult mode” would likely need to be accompanied by robust compliance systems in multiple jurisdictions before being deployed widely.

The announcement about ChatGPT’s delayed adult mode came as OpenAI faced internal controversy following the resignation of a senior executive linked to its robotics division.

Caitlin Kalinowski stepped down after raising concerns about the company’s partnership with the United States Department of Defense.

Kalinowski said she was troubled by the potential implications of AI technologies being used in areas such as mass surveillance or autonomous weapons systems.

“AI has an important role in national security,” she wrote in a statement on social media platform X. “But surveillance of Americans without judicial oversight and lethal autonomy without human authorisation are lines that deserved more deliberation than they got.”

She emphasised that her concerns related primarily to the speed with which the deal had been announced rather than the concept of national security collaboration itself.

“These are governance concerns first and foremost,” she said. “Issues this significant require clearly defined guardrails before agreements are announced.”

In response, OpenAI said it would update the terms of its defence agreement to ensure that its technology cannot be used for mass domestic surveillance or fully autonomous weapons systems.

A company spokesperson said the partnership was intended to support responsible national-security applications of AI while maintaining clear ethical boundaries.

“We believe our agreement with the Pentagon creates a workable path for responsible national security uses of AI while making clear our red lines: no domestic surveillance and no autonomous weapons,” the spokesperson said.

OpenAI added that it would continue engaging with employees, policymakers and civil society groups to ensure its technology is deployed responsibly.

The delay of ChatGPT’s adult mode reflects the broader challenge facing AI companies as they attempt to balance technological innovation, safety safeguards and regulatory compliance.

As generative AI tools become more widely used for everything from work productivity to creative expression, companies are increasingly under pressure to introduce new features carefully and responsibly.

For OpenAI, the immediate focus appears to be ensuring that ChatGPT’s core intelligence and usability continue to improve — a strategy the company believes will have a greater impact on its hundreds of millions of users than expanding the range of content the chatbot can produce.

Whether adult mode eventually launches may depend on how effectively OpenAI can implement reliable age verification and content moderation systems — a complex technical and legal challenge that is still evolving alongside the rapidly advancing capabilities of artificial intelligence.

Read more:
OpenAI delays ‘adult mode’ for ChatGPT to focus on higher priorities

Circular tech start-up Kibu has secured an investment offer from entrepreneurs Peter Jones and Jenna Meek following a televised pitch on Dragons’ Den, putting repairable children’s electronics firmly in the national spotlight.

The award-winning brand, which produces modular, repairable headphones for children, appeared on the long-running BBC programme represented by co-founder and chief executive Sam Beaney. Kibu’s pitch focused on its mission to redesign children’s consumer electronics around circular principles, prioritising disassembly, repair and customisation over disposal.

Founded through a collaboration between London-based design studio Morrama, advanced manufacturing partner Batch.Works and Beaney, Kibu first launched via a successful Kickstarter campaign. Since then, the company has transitioned from prototype to scalable commercial product, positioning itself as a challenger brand in a sector dominated by low-cost, disposable devices.

Kibu’s headphones are built with modular components that can be taken apart and reassembled by children. Individual parts can be replaced in minutes, extending product lifespan and reducing electronic waste. The design also allows for aesthetic customisation, enabling users to change colours and update components as preferences evolve.

The brand has already received international recognition for innovation and sustainability, tapping into growing parental demand for durable, repairable products in an era of heightened environmental awareness.

Speaking during the broadcast, Jones praised the concept and offered backing, citing his own early experience building and selling computers as a teenager. Meek also expressed interest in supporting the venture.

Beaney told the Dragons that empowering children to build and repair their own technology shifts their relationship with ownership and value. “When a child builds something themselves, it changes how they feel about it. When they learn they can fix what they’ve made, it changes how they see everything they own,” he said.

Jo Barnard, founder and creative director of Morrama, described the brand as a blueprint for futureproof electronics. By combining onshored manufacturing with agile supply chains, she argued, Kibu could unlock wider opportunities across children’s consumer technology.

Julien Vaissieres, chief executive of Batch.Works, said the project demonstrated how manufacturing can be structured to reduce waste while maintaining commercial viability. As both a founder and a parent, he said, the appeal lay in giving children agency over the products they use daily.

Now in its 23rd series, Dragons’ Den remains one of the UK’s most visible entrepreneurial platforms, attracting around three million viewers per episode on BBC One. For Kibu, the appearance offers both capital and brand recognition at a pivotal growth stage.

With investor backing now on the table, Kibu plans to scale distribution while continuing to develop its circular design ethos. The company believes its repair-first approach could extend beyond headphones into a broader range of children’s electronics, an industry segment increasingly scrutinised for its environmental footprint.

As sustainability pressures intensify and right-to-repair legislation gains momentum across global markets, Kibu’s model may offer an early glimpse of how future consumer electronics for children could be designed, manufactured and owned.

Read more:
Kibu secures investment offer from Peter Jones and Jenna Meek after Dragons’ Den pitch

OpenAI is to expand its London research centre to become its largest hub outside the United States, setting the stage for an intensified battle with Google DeepMind for top artificial intelligence talent in the UK capital.

The developer of ChatGPT said it would significantly increase the size of its London operation, which currently employs around 30 researchers, although it stopped short of disclosing specific headcount targets or investment figures. The move represents a strategic deepening of its UK presence at a time when competition for elite AI engineers has become one of the fiercest recruitment wars in global technology.

OpenAI, whose European headquarters remain in Dublin, said London offered a “unique concentration of world-class talent across machine learning and the sciences” alongside a strong culture of interdisciplinary collaboration.

The expansion is widely seen as a direct challenge to DeepMind, which employs about 2,000 staff in the UK and has long dominated Britain’s AI research ecosystem.

Mark Chen, OpenAI’s chief research officer, acknowledged that the company had already recruited staff from DeepMind and expected to continue doing so. He said OpenAI’s appeal lay partly in its culture.

“We are famously a bottom-up lab,” Chen said. “We let researchers pursue their lines of research and turn those into company-level bets.” By contrast, he suggested, Google’s approach could be “slightly more top-down”.

The contest for AI talent has driven compensation to extraordinary levels. Senior engineers at major AI labs can command packages worth well over £1m, often made up of salary, bonuses and equity. In the United States, reports have emerged of multi-million-dollar offers as firms scramble to secure leading researchers.

As a private company, OpenAI can offer equity stakes that may rise significantly in value if the firm eventually lists publicly. It has also facilitated secondary share sales, allowing employees to monetise part of their holdings — creating a powerful recruitment incentive.

Chen said compensation would remain “very competitive”, adding: “AI talent is very valuable and we need to be competitive everywhere.”

The expansion was welcomed by UK political leaders eager to position Britain as a global AI powerhouse.

Technology secretary Liz Kendall described the move as “a huge vote of confidence in the UK’s world-leading position at the cutting edge of AI research”.

London mayor Sadiq Khan said he was “delighted that OpenAI is anchoring its major new research hub here”, arguing that the capital’s academic institutions and tech ecosystem made it a natural home for the next wave of AI innovation.

The announcement comes as the UK government seeks to attract high-growth technology firms as part of its broader economic strategy, positioning AI as a central driver of productivity and competitiveness.

OpenAI’s expansion follows internal warnings from its chief executive, Sam Altman, that the company faced mounting competition from rivals including Google and Anthropic. Altman has previously described the race in advanced AI development as a “code red” moment for the firm.

Chen said recent advances in so-called AI agents — autonomous software capable of executing tasks with limited supervision — marked a significant inflection point for the industry.

“Something is happening in AI that feels like a step change,” he said. “We’ve reached a level where we can rely on agents and use them in real-world workflows.”

He described how researchers can now delegate experiment execution to AI systems, returning to interpret results and refine hypotheses. This, he suggested, would increasingly reshape not only research roles but also broader “analyst-style” professions.

However, Chen cautioned that such systems remain dependent on human oversight and design. “Agents cannot ideate and come up with the experimental design itself,” he said.

As AI capabilities accelerate, public anxiety about job displacement and social impact has grown. Recent essays questioning the pace and implications of AI development have circulated widely, contributing to volatility in technology markets.

Chen acknowledged that “external perception of AI has shifted in a more negative direction” but argued that many practical applications — particularly in productivity and research — remain underappreciated.

“There are many positive uses of agents,” he said. “That’s something we as an industry need to underscore.”

With OpenAI committing to scale up in London, the capital is poised to become an even more intense battleground in the global race to dominate advanced AI — with research talent, equity incentives and cultural positioning now as important as computing power itself.

Read more:
OpenAI to make London its largest research hub outside US

The UK’s data protection watchdog has fined Reddit £14.47m after finding serious failings in how the platform handled children’s personal information.

The penalty, issued by the Information Commissioner’s Office (ICO), follows an investigation that concluded Reddit had failed to implement robust age assurance mechanisms and did not have a lawful basis for processing the data of children under 13.

Under UK data protection law, children’s information must be given special protection. The ICO said Reddit did not have effective systems in place to verify users’ ages until July 2025, despite its terms of service prohibiting under-13s from accessing the platform.

The regulator also found that Reddit failed to carry out a data protection impact assessment (DPIA) addressing risks to children until January 2025, even though users aged 13 to 18 were permitted to join.

John Edwards, the UK Information Commissioner, described the failings as unacceptable. “Children under 13 had their personal information collected and used in ways they could not understand, consent to or control,” he said. “Relying on users to declare their age themselves is not enough when children may be at risk.”

In July 2025, Reddit introduced new measures including age verification for access to mature content and requiring users to declare their age at account creation. However, the ICO has warned that self-declaration alone presents risks, as it can be easily bypassed.

The regulator said it would continue monitoring Reddit’s approach as part of wider enforcement activity focused on online platforms that rely primarily on self-declared ages.

The fine takes into account the number of children potentially affected, the duration of the failings and Reddit’s global turnover.

The ICO’s action follows its ongoing supervision of platforms under the UK’s Age Appropriate Design Code, also known as the Children’s Code, which sets out standards for services likely to be accessed by under-18s.

The regulator has said safeguarding children’s privacy online remains a priority and confirmed it will continue working closely with Ofcom, which enforces the Online Safety Act, to ensure coordinated oversight of digital platforms.

The decision underscores intensifying scrutiny of tech companies operating in the UK, particularly around age verification and the lawful processing of children’s data.

Read more:
ICO fines Reddit £14.47m over children’s data protection failures

Improved mobile connectivity could help create 49,000 new businesses across the UK and add £6.6bn a year to the economy within a decade, according to research commissioned by VodafoneThree.

The modelling, carried out by consultancy WPI Strategy, suggests that stronger and more reliable mobile coverage would unlock entrepreneurship in underserved areas, driving long-term economic growth by 2036.

The findings come as VodafoneThree announced it had removed 16,500 square kilometres of mobile “not spots” by deploying Multi Operator Core Network (MOCN) technology across more than 8,000 sites nationwide. The technology allows Vodafone and Three customers to connect to the strongest available signal at no extra cost.

The upgrade forms part of the company’s £11bn investment programme, which aims to deliver 99 per cent 5G Standalone population coverage by 2030, rising to 99.96 per cent by 2034.

An independent survey of 2,000 people, including existing and aspiring business owners, found that 62 per cent of would-be founders said unreliable mobile connectivity had prevented them from starting a business in their local area.

A third said better signal would make their area more attractive for launching a company, while 26 per cent said it would directly increase their likelihood of setting up a business locally.

The research echoes findings from the Department for Science, Innovation and Technology that dependable mobile connectivity boosts entrepreneurship and business performance, particularly in rural areas.

Nick Gliddon, business director at VodafoneThree, said: “When connectivity improves, entrepreneurship follows. Strong and reliable networks help start-ups win customers, build reputation and grow steadily.”

The North West of England is forecast to be among the biggest beneficiaries, with improved coverage potentially supporting nearly 6,000 new firms and adding an estimated £807m annually to the regional economy within 10 years. The South East could see around 5,800 new businesses, contributing £784m.

Even London, often assumed to be well served, stands to gain. The research suggests enhanced connectivity in the capital could enable more than 14,000 new businesses and contribute £1.9bn to the economy. Westminster alone represents the largest single opportunity, with additional gains projected in boroughs including Camden, Hackney and Islington.

Elsewhere, Wales could see over 1,000 new firms created, worth £136m annually, while Scotland could gain more than 2,100 businesses contributing £291m.

Connectivity challenges are already shaping business decisions. Two in five founders surveyed said they had relocated to start their company, citing poor signal, limited customer bases and restricted access to talent.

Six in 10 entrepreneurs said they rely on mobile connectivity to run their operations, while nearly nine in 10 reported having experienced outages that disrupted trading.

Tina McKenzie, policy chair at the Federation of Small Businesses, said consistent 5G rollout remained essential. “If we want more people to take the leap into starting their own business, they need reliable connectivity to make it possible,” she said.

Telecoms minister Liz Lloyd added that the government was working with network operators to improve coverage and support enterprise ambitions across the country.

With digital infrastructure increasingly central to modern commerce, from payments and marketing to logistics and customer service, VodafoneThree argues that closing connectivity gaps could be a critical lever for unlocking the UK’s next wave of entrepreneurial growth.

Read more:
Improved mobile coverage could unlock 49,000 new UK businesses, VodafoneThree says

A Manchester business owner was left facing a £42,000 mobile phone bill after his daughter streamed TikTok on holiday in Morocco, a charge he said nearly pushed his small company into financial crisis.

Andrew Alty, who runs a curtains business, discovered the scale of the bill while on a family trip to Marrakech after receiving an initial £22,000 invoice from O2. A second bill for around £20,000 followed shortly after their return to the UK.

The charges stemmed from data roaming outside Europe, where the UK’s previous EU-wide free roaming arrangements no longer apply. Mr Alty had taken out a mobile contract via Currys for his small business, unaware that the agreement included a clause opting out of a “rest-of-world” data cap.

His daughter had spent around eight hours using TikTok during the trip. With no cap in place, data charges mounted rapidly, amounting to more than £5,000 per hour of usage.

“There’s no way they should be able to charge that,” Mr Alty told The Telegraph. “They made no effort to inform us and just allowed the charges to accrue. I don’t understand how they expect any small business to pay that sort of bill.”

Initially suspecting fraud or a technical glitch, Mr Alty attempted to contact O2 while still abroad but was unable to resolve the issue. It was only after returning home that the family understood the source of the charges.

Following weeks of complaints, both Currys and O2 agreed to waive the bill in full.

According to Ofcom data covering July to September 2025, O2 received among the highest levels of complaints per 100,000 customers, alongside Sky Mobile and Three. Almost a third related to complaints handling.

Mr Alty escalated his case to the Financial Ombudsman Service, arguing that the opt-out clause on data caps had not been clearly explained. However, the ombudsman ruled that contract explanations fell under Currys’ responsibility, not O2’s. The FOS does not adjudicate disputes with network providers directly.

An O2 spokesperson said the matter had been resolved following Currys’ internal review, with all charges waived “given the scale and circumstances”.

The case highlights the potential financial risks of using mobile data outside Europe without a roaming cap, particularly on business contracts where standard consumer protections may differ. While most major networks offer optional caps to limit overseas data costs, customers must ensure these safeguards are activated, or risk bills running into the tens of thousands.

Read more:
£42,000 roaming bill nearly bankrupts family firm after TikTok use abroad

The owners of Virgin Media O2 have agreed a £2bn takeover of challenger fibre network Netomnia, marking a significant step towards consolidation in Britain’s crowded broadband market.

Liberty Global and Telefónica, alongside InfraVia Capital through their Nexfibre joint venture, will acquire Netomnia, currently the UK’s second-largest alternative network provider.

The deal will expand Nexfibre’s footprint to around 8 million households by the end of next year. Combined with Virgin Media O2’s existing infrastructure, the enlarged network will cover approximately 20 million premises and serve about 6.2 million customers.

That scale brings it close to Openreach, the network arm of BT Group, which has passed just over 21 million premises with full fibre.

Shares in BT fell 2.5 per cent following news of the acquisition.

Founded in 2019, Netomnia is one of dozens of “altnets” that emerged to challenge the dominance of Openreach and Virgin Media O2. However, many smaller fibre operators have paused expansion amid higher borrowing costs and weaker-than-expected customer take-up.

Rajiv Datta, chief executive of Nexfibre, said the enlarged group would offer greater scale to wholesale partners, including Sky, which recently began using CityFibre’s network in addition to Openreach.

The transaction saw Virgin Media O2 beat CityFibre, backed by Goldman Sachs, which has previously positioned itself as a natural consolidator of the fragmented sector.

Simon Holden, chief executive of CityFibre, criticised the move, warning it risked recreating an “ineffective duopoly” between BT and Virgin Media O2 and calling on the Competition and Markets Authority to scrutinise the overlap.

The acquisition will be financed with £850m in equity from InfraVia and £150m from Liberty Global and Telefónica, alongside a £2.7bn debt facility to fund both the purchase and further network expansion.

The deal comes as Virgin Media O2 continues to face customer losses, shedding 18,000 broadband subscribers and 165,000 mobile customers in the latest quarter.

Separately, Liberty Global has agreed to pay €1bn to Vodafone for its 50 per cent stake in VodafoneZiggo, the Dutch joint venture. Liberty plans to merge VodafoneZiggo with its Belgian unit Telenet and spin off the combined entity, Ziggo Group, via a listing in Amsterdam next year.

The Netomnia acquisition signals that consolidation in the UK fibre market, long expected as funding tightens and competition intensifies, is now gathering pace, potentially reshaping the balance of power in Britain’s broadband industry.

Read more:
Virgin Media O2 owners strike £2bn deal for Netomnia in fibre consolidation push

Alphabet has unveiled plans to spend between $175bn and $185bn this year, sharply exceeding Wall Street expectations as it intensifies its push in the global artificial intelligence race.

The capital expenditure target is well above analysts’ average forecast of about $115bn, according to LSEG data, and marks another escalation in spending among the world’s technology hyperscalers.

The announcement came alongside strong fourth-quarter results. Revenue rose 18 per cent year-on-year to $113.8bn, narrowly ahead of forecasts of $111.3bn. Net income climbed 30 per cent to $34.5bn, comfortably beating expectations of $31.9bn.

Despite the earnings beat, Alphabet shares slipped 1.4 per cent in after-hours trading, reflecting investor unease over the scale of spending commitments.

Under chief executive Sundar Pichai, Alphabet has repositioned itself as a leading force in AI after earlier concerns that start-ups such as OpenAI might disrupt its core search business.

Google’s Gemini model has become a central pillar of its strategy, with the Gemini AI assistant app exceeding 650 million monthly users in November. Its AI Overviews feature within search has reached more than 2 billion monthly users.

The company is also investing heavily in custom AI chips and data centre infrastructure, which investors hope will drive future growth.

Last month, Google secured a high-profile partnership with Apple to power an upgraded version of Siri with Gemini models, opening access to Apple’s installed base of more than 2.5 billion devices.

Nikhil Lai, principal analyst at Forrester, said the results demonstrated resilience in Alphabet’s core advertising business. “Record ad revenue signals sustained momentum in search and solid performance from YouTube,” he said, noting that YouTube’s scale now exceeds that of Netflix.

Alphabet’s shares have surged over the past year, rising more than 64 per cent and pushing its market capitalisation above $4tn — second only to Nvidia, valued at around $4.3tn.

However, wider market sentiment towards AI stocks has turned more cautious. Last week, Microsoft reported slower cloud growth, prompting a sell-off amid concerns about the sustainability of heavy AI investment. While Meta reassured investors with upbeat revenue guidance, other names struggled.

The S&P 500 and Nasdaq both declined as investors reassessed lofty valuations. Shares in Advanced Micro Devices fell sharply after a weak revenue outlook, while Palantir also dropped on AI spending concerns.

Jed Ellerbroek, portfolio manager at Argent Capital, said the scale of AI infrastructure build-out was unprecedented. “The market is having a hard time knowing where to price these stocks and what the future looks like,” he said. “There’s growing scepticism about whether the rally has peaked.”

For Alphabet, the strategy is clear: double down on infrastructure to secure long-term AI leadership. Whether investors remain willing to fund that ambition at such scale will depend on how quickly those vast capital commitments translate into durable returns.

Read more:
Alphabet ramps up AI spending with up to $185bn capital plan

UK semiconductor start-up Fractile has announced a £100 million expansion of its British operations, scaling up in London and Bristol as ministers intensify calls for greater domestic ownership of critical artificial intelligence technology.

The investment, to be deployed over the next three years, will fund a new industrial hardware engineering facility in Bristol, alongside the expansion of Fractile’s existing UK sites and a significant increase in its domestic workforce.

The company is focused on developing AI chips optimised for inference, the stage at which large language models generate outputs, an area of growing strategic importance as demand for real-time AI applications accelerates.

Engineers at the new Bristol facility will work on integrating Fractile’s chips into full AI systems and will operate a specialist software testing lab, allowing the company to develop and validate hardware and software in tandem.

The announcement comes as Kanishka Narayan, the government’s AI minister, prepares to urge Britain’s technology founders and investors to “embrace risk” and back home-grown innovation in a speech to the UK’s AI sector. He is expected to stress that British ownership of foundational technologies will be critical if the UK is to shape the future direction of AI.

Founded in 2022, Fractile is developing in-memory computing chips designed to run powerful AI models faster and with significantly lower energy consumption than conventional hardware. The market for AI inference chips is currently dominated by Nvidia, but is increasingly attracting start-ups and hyperscalers seeking more efficient and lower-cost alternatives.

Fractile is backed by the NATO Innovation Fund and has raised more than $35 million (£25.5 million) to date. The company says its technology could dramatically reduce both the cost and power required to run large AI models, an increasingly pressing constraint as data centre demand surges globally.

The expansion is being viewed as a vote of confidence in the UK’s ambitions to build a domestic AI hardware ecosystem, alongside continued investment in software, data and infrastructure. Ministers have identified “sovereign” computing capacity as a national priority, amid rising concerns over supply chains, ownership and national security.

Fractile said the £100 million commitment underlined its long-term intention to build and scale advanced semiconductor hardware on home soil, as scrutiny intensifies across the UK tech sector over who controls critical digital infrastructure.

The move follows a strong year for government-backed AI initiatives, with tens of billions of pounds of private capital pledged to UK AI projects and thousands of jobs expected to be created under the government’s year-old AI Opportunities Action Plan.

Read more:
Fractile commits £100m UK expansion as it ramps up AI chip development

Apple and Google have agreed to make changes to their UK app stores following intervention by the country’s competition watchdog, after it concluded the two firms hold an “effective duopoly” over the sector.

The Competition and Markets Authority (CMA) said the tech giants have committed to a series of measures designed to improve transparency and competition. These include pledges not to give preferential treatment to their own apps and to be clearer about how third-party apps are reviewed and approved for sale.

The commitments come seven months after the CMA warned that Apple and Google’s dominance of mobile app distribution in the UK was stifling competition. In October 2025, the regulator formally designated both companies’ app stores as having “strategic market status”, giving it enhanced powers to demand changes under the UK’s new digital competition regime.

Sarah Cardell, the CMA’s chief executive, said the agreements marked an important milestone. “These proposed commitments will boost the UK’s app economy and are the first of many measures,” she said. “The ability to secure immediate commitments from Apple and Google reflects the unique flexibility of the UK’s digital markets competition regime and offers a practical route to swiftly address the concerns we’ve identified.”

As part of the deal, both companies have also agreed not to use data collected from third-party app developers in ways the regulator considers unfair. Cardell described the changes as “important first steps”, adding that the CMA would continue working with the companies on further remedies.

The watchdog said it would “closely monitor” implementation and would not hesitate to impose legally binding requirements if the commitments were not honoured.

Both companies welcomed the outcome. Apple said it faced “fierce competition in every market where we operate” and that it was committed to delivering the best possible products and user experience. Google said it believed its existing practices on the Play Store were already fair and transparent, but added that it “welcomes the opportunity to resolve the CMA’s concerns collaboratively”.

Analysts cautioned that the agreement may not be the final word. Paolo Pescatore, a technology analyst, described the move as a “pragmatic first step” but said some critics would view it as tackling “low-hanging fruit”. “There will inevitably be calls for a tougher clampdown from some quarters,” he said.

The CMA said the UK app economy is the largest in Europe by revenue and number of developers, generating an estimated 1.5 per cent of UK GDP and supporting around 400,000 jobs.

Both Apple and Google have previously warned against the UK adopting rules similar to those in the European Union, where large online platforms designated as “gatekeepers” face sweeping obligations. Apple has already been forced in the EU to introduce changes such as offering users a choice of default browser, and has argued that some requirements undermine privacy and security.

Apple said the UK commitments reflected its “constructive engagement” with the CMA and a more pragmatic approach to regulation — but the regulator has made clear that further intervention remains on the table if competition concerns persist.

Read more:
Apple and Google agree UK app store changes after ‘effective duopoly’ ruling

Training workers to use artificial intelligence will be “critical” to managing disruption in the UK labour market, according to Andrew Bailey, who said there were already signs that AI was reshaping careers and hiring patterns.

Speaking at a conference in Saudi Arabia on Sunday, Bailey said the long-term impact of AI on employment remained “highly uncertain”, but warned that early indicators pointed to meaningful change.

“In the UK, in the last three years, new online vacancies in the most AI-exposed roles have decreased by more than twice as much as in the least exposed group,” he said.

“On the positive side, however, there has been a significant increase in new tasks, such as integrating AI tools into firms’ workflow processes.”

Bailey cautioned against drawing simplistic conclusions about the effect of AI on jobs, stressing that education and reskilling would be central to ensuring workers were not left behind. “Education and training in AI skills will be critical,” he said. “We shouldn’t resort to oversimplified conclusions on the employment effects.”

His comments came at the end of a volatile week for global markets, during which renewed anxiety over artificial intelligence wiped more than $1 trillion off the combined value of the world’s largest technology and software companies.

Investor nerves were rattled in part by new product launches from Anthropic, one of the world’s leading AI developers. The company unveiled tools aimed at automating legal work such as contract review, alongside its latest Claude Opus 4.6 model, which is capable of analysing complex information and producing presentations and spreadsheets.

The developments fuelled fears about job displacement and business model disruption, triggering sharp share price falls among UK-listed companies seen as highly exposed to AI. These included RELX, London Stock Exchange Group, and Sage.

At the same time, concerns grew that enthusiasm for AI may have run ahead of reality in the US technology sector. Amazon, Alphabet, Meta and Microsoft have collectively committed to spending around $660 billion this year on data centres and advanced computer chips to support AI development.

Fears that such vast capital investment may not deliver sufficient returns have weighed on share prices, adding to wider market turbulence. The pullback follows years of strong gains in US technology stocks, driven by investor optimism about AI-led productivity gains, optimism that has also raised concerns about a potential bubble.

Bailey said there were signs of “fear of missing out” in markets, reinforced by claims that AI represents a structural break from previous technology cycles. “We have seen arguments along the lines of ‘this time is different’, for instance because of the expected productivity benefits of AI,” he said.

He warned that this narrative risked complacency among investors and policymakers alike. “Expectations of AI-driven productivity gains could be disappointed,” he said.

Despite the caution, Bailey struck a broadly optimistic note on the long-term economic potential of AI and robotics. He said he believed the technologies could boost productivity and growth by automating repetitive tasks and creating entirely new types of work.

However, he added that the transition would not be painless. “Some industries might shrink, others grow, and affected workers will need to retrain to adapt their skills,” he said, underlining once again that investment in training would be decisive in shaping the future of the UK jobs market.

Read more:
Andrew Bailey warns AI training is critical to future of UK jobs

The UK’s data protection watchdog has fined the owner of image-sharing platform Imgur nearly £250,000 after finding serious failures in how the site handled children’s personal data.

The Information Commissioner’s Office (ICO) has imposed a £247,590 penalty on MediaLab.AI, Inc, concluding that the company allowed children to access Imgur for years without putting in place even basic age-checking safeguards required under UK data protection law.

Following a long-running investigation, the regulator found that MediaLab failed to establish the age of Imgur users, processed the personal data of children under 13 without parental consent, and did not carry out a data protection impact assessment to identify or mitigate risks to younger users.

Because Imgur had no effective way of determining who was using the platform, children were exposed to potentially harmful content, including material relating to eating disorders, antisemitism, homophobia, and sexually explicit or violent imagery. The ICO said personal data was being used to shape content recommendations without any protections appropriate for children.

John Edwards, the UK Information Commissioner, said the company had failed in its legal duty to protect young users. He said MediaLab had allowed children to use Imgur without effective age checks while collecting and processing their data, exposing them to serious risk.

He added that age assurance plays a crucial role in protecting children’s personal information and preventing it from being used in ways that may cause harm, such as recommending age-inappropriate content. Companies that ignore the fact children use their services, he warned, should expect enforcement action.

The ICO’s investigation covered a four-year period between September 2021 and September 2025, during which MediaLab was found to have breached the UK General Data Protection Regulation. Under UK law, online services can only rely on consent as a lawful basis for processing data relating to children under 13 if that consent is given by a parent or carer.

Although Imgur’s terms stated that children under 13 required parental supervision, the ICO found that MediaLab had no mechanisms in place to enforce this or obtain parental consent.

In setting the penalty, the regulator took into account the length of time the breaches occurred, the number of children affected, the level of potential harm, and MediaLab’s global turnover. The ICO also noted that MediaLab accepted its provisional findings and committed to implementing appropriate safeguards should Imgur resume processing children’s data in the UK.

The watchdog said further regulatory action could follow if those commitments are not met.

The fine forms part of the ICO’s broader push to improve how digital platforms protect children’s personal information online. UK data protection law gives children enhanced protections, reinforced through the Children’s Code, also known as the Age Appropriate Design Code, which sets clear expectations for online services likely to be accessed by under-18s.

The ICO has stressed that platforms must either apply the Children’s Code protections to all users or implement proportionate and robust age assurance measures to tailor safeguards appropriately.

Read more:
ICO fines Imgur owner £247k over children’s data failures

The UK Government has announced a £36 million investment to expand access to advanced artificial intelligence computing, backing a major upgrade of the University of Cambridge’s DAWN supercomputer.

Ministers say the move will give British researchers and startups free access to high-performance AI computing power that is typically dominated by global technology giants, helping to level the playing field for smaller teams working on public-interest innovation.

The funding will increase DAWN’s capacity sixfold within months, allowing hundreds more research projects to run alongside the 350 already using the system. The government says the upgraded supercomputer will support breakthroughs in personalised cancer treatment, climate modelling and earlier disease detection in primary care.

According to the Department for Science, Innovation and Technology, British scientists are already using DAWN to identify which parts of a tumour the immune system is most likely to attack, refine flood prediction models for local authorities and develop AI tools that could help GPs diagnose conditions earlier.

AI minister Kanishka Narayan said the investment addressed a longstanding barrier for UK innovation.

“The UK is home to world-class AI talent, but too often our most ambitious researchers and startups have been held back by a lack of access to computing power,” he said. “This investment gives British innovators the tools they need to compete with the biggest players and build AI that delivers real benefits, from healthcare to climate resilience.”

While the announcement has been welcomed as a practical step to support domestic research, industry figures are divided over whether the scale of funding matches the global reality of AI investment.

Colette Mason, author and AI consultant at Clever Clogs AI, said the value of the investment depends on how its outcomes are governed.

“£36 million is good value if it shortens diagnosis timelines, improves flood planning or strengthens public services in ways people can see,” she said. “It’s poor value if the upside ends up locked into private intellectual property or acquisitions that move the benefit elsewhere. Public investment should come with public conditions.”

Others were more sceptical. David Belle, founder of Fink Money, contrasted the funding with levels of investment seen elsewhere.

“In global terms, £36 million is a tiny sum,” he said. “The US has committed billions to non-defence AI research. There’s a risk this money disappears into planning and consultation rather than delivery.”

However, Rohit Parmar-Mistry, founder of Pattrn Data, argued that the investment should be judged on focus rather than scale.

“In the global AI arms race, £36 million is a rounding error. Silicon Valley spends that before breakfast,” he said. “But the UK doesn’t need to out-spend Big Tech, it needs to out-think it. Expanding access to compute for British researchers is a smart move, provided the public retains a stake in what gets built.”

The government says the DAWN expansion forms part of its wider AI strategy to improve access to compute, accelerate applied research and ensure public-sector challenges, from healthcare to climate adaptation, are not sidelined by commercial priorities.

Read more:
UK invests £36m in AI supercomputer to boost research and startup innovation

Pinterest has dismissed two engineers after they created and shared a software tool that identified colleagues who had been made redundant during a recent round of job cuts, according to reports.

The digital pinboard company announced earlier this month that it would cut about 15 per cent of its workforce, roughly 700 roles, as chief executive Bill Ready said the business was “doubling down on an AI-forward approach”. Pinterest did not disclose which teams would be affected by the reductions.

Following the announcement, two engineers wrote custom scripts that accessed internal systems to flag when employee accounts were deactivated, effectively revealing the names and locations of staff who had lost their jobs. The information was then shared more widely, prompting the company to take disciplinary action.

A Pinterest spokesperson said the engineers had “improperly accessed confidential company information” and described the actions as a clear breach of company policy and a violation of affected employees’ privacy. It remains unclear whether the data was shared solely with colleagues inside the business or beyond the company.

The scripts targeted internal communication and access tools, according to the BBC, citing a source familiar with the incident. The code reportedly triggered alerts when employee names were removed from internal systems.

Pinterest has been ramping up investment in artificial intelligence to improve personalisation for users and automate tools for advertisers. However, investor confidence has been shaken, with shares down more than 20 per cent this year as markets weigh the competitive threat posed by newer and more advanced AI platforms.

Ready told staff in an internal meeting that while debate and dissent were healthy, employees who fundamentally disagreed with the company’s direction should consider their future elsewhere, according to CNBC, which first reported the firings.

The incident comes amid a broader wave of job losses across the tech sector as companies restructure around AI. Last week, Amazon announced a further 16,000 redundancies worldwide, while Meta said it would cut more than 1,000 roles from its Reality Labs division. Design software maker Autodesk has also confirmed plans to shed around 1,000 jobs this month.

Read more:
Pinterest sacks engineers after internal tool exposed laid-off staff during AI-driven cuts