Engineers as well as designers from Axiom Space and Prada pulled back the curtain last weekend in New York on the inner layer that will sit closest to astronauts during future lunar surface work. The garment forms a key piece of the AxEMU spacesuit developed for NASA’s Artemis program. Astronauts step into this form-fitting piece first. Light gray fabric stretches across the body in a streamlined silhouette while clear tubing traces deliberate paths over the torso, arms, and legs. A single red stripe runs down one sleeve as a quiet nod to Prada’s activewear roots.

Water flows through the primary network of tubes, absorbing heat from the astronaut’s muscles. That warmed liquid then travels to the portable life support system strapped to the astronaut’s back, where it releases its heat into space. Meanwhile, an entirely distinct system supplies new oxygen to the helmet and collects exhaled carbon dioxide, which is then cleaned and reused. In fact, added redundancy adds another layer of protection: a second, fully operational cooling loop runs parallel to the first. If the main system fails, the backup system takes over seamlessly, ensuring that temperature control is never disrupted during one of these excursions.

LEGO Technic NASA Artemis Space Launch System Rocket Building Toy for Boys & Girls – STEM Learning…

• BUILD AN OFFICIAL NASA ROCKET – Kids prepare to explore outer space with the LEGO Technic NASA Artemis Space Launch System Rocket (42221) building…
• 3-STAGE ROCKET SEPARATION – Young builders can turn the hand crank to watch the rocket separate in 3 distinct stages: solid rocket boosters, core…
• STEM BUILDING TOY FOR KIDS – This educational rocket kit was created in collaboration with NASA and ESA to showcase the authentic system that will…

Prada contributed decades of expertise in high-end textiles, unique knitting processes, and precision garment manufacturing to this project. Axiom offered aerospace expertise, life support integration, and 3D modeling tools that designed each tube route to improve cooling efficiency while allowing the wearer to move freely. The two teams worked back and forth through several different models, determining which materials performed best in certain scenarios and on different body types. A set of connectors at the waist connects the garment’s tubes directly to the rest of the suit’s systems. The entire design is suitable for eight-hour spacewalks and is built to be reused over long-duration missions, thanks to specific fibers selected with Prada’s involvement.

This suit’s inner layer outperforms the spacesuits currently deployed on the International Space Station. The earlier designs just did not have that level of built-in backup cooling, and the degree of customisation available with these newer suits is a huge advantage. Materials were also chosen to solve specific issues connected with the lunar environment, such as plasma interactions, which prior suits were never designed to handle. Axiom Space was awarded a contract from NASA to develop these suits several years ago. The qualification gear is presently being tested on the ground, with an in-space demonstration scheduled for next year, and the company is on track to support crewed trips to the lunar surface during the Artemis IV mission later this decade.
[Source]

After being requested for years, Capcom is finally bringing one of its Resident Evil titles back to life. Resident Evil Veronica has been announced for a 2027 release, bringing back Resident Evil Code: Veronica–all with modernized gameplay, a reimagined storyline, and of course, next-gen visuals. The game is headed to PlayStation 5, Xbox Series X/S, Nintendo Switch 2, and PC via Steam.

Claire Redfield takes the lead again

Just like before, Resident Evil Veronica kicks off after the events of Resident Evil 2. A couple of months after the Raccoon City disaster, Claire travels to France in search of her brother, Chris Redfield. However, the reunion goes awry with Umbrella’s special forces capturing her and sending her to Rockfort Island, which soon becomes another biological disaster zone.

This was the exact setup for Code: Veronica, which is also what set it apart from the other mainline entries in the popular franchise. It was never numbered, which made it easier for casual players to skip, but it is far from a throwaway side entry. It continues Claire and Chris Redfield’s story, leans heavily into Umbrella’s lingering mess, and helps bridge the older survival-horror era with the more dramatic tone the franchise would later embrace.

Why fans have been waiting for this remake

Resident Evil Code: Veronica has long sat in an odd place within the series, overlooked by some because it was not a numbered entry, but kept alive by a dedicated fanbase that never stopped asking for its return. With more older titles getting fresh remakes, it made sense for Veronica to be next–and Capcom did not disappoint. Resident Evil Veronica preserves the essence of the 2000 game while adding modern gameplay and a reworked story. 

The original Code: Veronica is beloved, but it is also one of the less forgiving classic Resident Evil games. Its old-school structure, harsh resource management, and dated controls made it memorable for the right and wrong reasons. So a modern remake could make it far more approachable without losing the nasty, isolated atmosphere that made Rockfort Island stand out. 

At the moment, Capcom has only given Resident Evil Veronica a broad 2027 release window. The Steam page does not list a price or exact launch date yet, and system requirements are still marked as TBD. Even with those missing details, this is a big one for Resident Evil fans. After remakes of Resident Evil 2, Resident Evil 3, and Resident Evil 4, Code: Veronica was the obvious unfinished business. Now, it is finally getting its turn. 

British Prime Minister Keir Starmer on Monday asked tech companies operating in the UK to prevent children from taking, sending and receiving nude images. If they fail to do so, the UK will change the law, he said.

“I am calling on tech companies operating in this country to introduce device controls that prevent children from sending and receiving sexually explicit images,” Starmer said in a speech at London Tech Week. “This is not an impossible challenge. These are some of the most innovative companies in the world and I believe they can solve it, but if they choose not to, then we will act.”

The British government is asking tech companies, including Apple and Google, to put in place device-level controls to prevent nude images of children within the next three months. It wants the technology to be available across both new and existing phones.

“Google is deeply committed to protecting children online. We are working constructively with UK partners to find effective, privacy-preserving solutions that deter the spread of harmful content while ensuring a safe digital environment for young people,” said a Google spokesperson in an emailed statement.

A representative for Apple didn’t immediately respond to a request for comment.

Starmer is also mulling new rules that would ban social media for children under the age of 16, similar to the ban already in place in Australia, according to the Times.

Many children, especially teenagers, take, share and receive naked photos without their parents being aware. This behavior is illegal and also puts children at risk of blackmail, bullying, sexual harassment or child exploitation. The UK is the first country to demand that tech companies help keep children safe by preventing them from taking nude photos and circulating them online.

In a statement supporting Starmer’s announcement on device controls, the UK’s National Crime Agency said the restrictions have the potential to stop some of the most serious forms of online child sexual abuse before they begin. 

“Many of the most serious cases begin with offenders coercing children into creating and sharing sexual images of themselves,” said Graeme Biggar, NCA director general. “Once those images exist, they can be used for blackmail, humiliation and repeated exploitation. Preventing children from taking, sharing or receiving nude images can stop abuse before it starts and so would be an important step forward.”

But not everyone welcomed Starmer’s comments. “Protecting children online is vital, but these are outrageous plans that will fail to address the underlying causes of online harm,” Silkie Carlo, director of rights group Big Brother Watch, said in a statement. “This will only result in population-wide ID checks for all of us to use our phones, tablets and laptops.”

Starmer’s request isn’t designed to prevent adults from taking and sending nudes, but attempts to make the internet safer for young people can also affect adults online. The UK’s Online Safety Act requires that many corners of the internet, including Reddit, verify the ages of people accessing their services. In the same way, it’s entirely possible that this request could result in people having to prove their identity in order to use a device without a child lock.

PwC’s Raymond Martin explores his day to day at the intersection of technology and healthcare.

“I lead technology-driven transformation for PwC’s health sector clients in Ireland, helping them improve patient care, expand access to services and run more efficient operations,” said Raymond Martin, a health technology director for the organisation.  

He explained that, more and more, healthcare teams are being asked to do more with the same resources, so targeted transformation using technology is often the single biggest lever for change. 

Here he discusses his day to day at PwC.

If there is such a thing, what is a typical day in your job like?

There really isn’t one and that’s a positive. The pace of change across technology, health and business over the last five years has been astonishing and AI has put it into overdrive in the last two.

My time tends to split three ways. First, overseeing live client delivery, technology roll-outs in acute hospitals, community teams and national programmes, often involving significant redesign of care pathways to enhance collection and use of data and promote data and process standardisation. Second, working with our clinical and operational experts to shape new services based on what clients are asking for and the trends we’re seeing globally. Third, helping our PwC team evolve and working out what the market shift means for our people, their skills and their careers.

In the more than 20 years you have worked in the health-tech space, how has the landscape evolved?

It’s night and day. 20 years ago, health-tech was dominated by back-office systems, heavy custom development, complex configuration projects and clinical systems that were largely paper-based or niche high tech. Delivery of any change took years.

When I started on Medicaid systems in the US in 2008, the average time from design to first release was four years and some releases might be a year apart. Five years later, working on the Patient Protection and Affordable Care Act in California, that had dropped to two years. The explosion of cloud infrastructure and SaaS platforms brought it down to under a year and Covid compressed it again to months. 

During Covid, for the New Mexico Department of Health, I led a team through mobilisation of a contact centre, CRM deployment and the onboarding of hundreds of staff, beginning operations within 30 days and adding most functionality within three months of that. The challenge for us is making sure governance and methodology keep pace so speed doesn’t introduce new risk. The other big shift is the democratisation of technology. 20 years ago, technology was a niche skill. Today our clients come prepared, fluent in the tools and often using AI to enhance their own awareness and position in conversations.

How is technology transforming the future health space?

It’s hard to find a corner of healthcare that technology won’t touch. The clinician-patient relationship will remain the human core, but everything around it is changing fast. Data is the first frontier. Healthcare generates extraordinary volumes of rich data, but it has historically been trapped in silos, useful inside one system or location, invisible everywhere else.

Joining that data across boundaries unlocks huge value, like better national and local service planning for a start. With better access to clinical data and the ability to augment it with lifestyle data from wearable devices and eating habits, patients can actively manage their own health as well as their care journeys. The most obvious advantage is it gives clinicians a fuller picture of the person in front of them.

Beyond the clinical interaction, cloud platforms and AI are reshaping access to care. Virtual consultations, remote monitoring and hospital-at-home are becoming normal. AI assistants will help people build healthier habits and manage conditions preventatively. Genetic screening will drive tailored therapies into the mainstream and AI-accelerated trials will shorten the path from discovery to approval. And that’s just the start.

What are some of the challenges and how are they managed?

The pace of technology change is now stress-testing the governance and procurement models organisations rely on. They simply weren’t built for this rate of arrival.

Healthcare and public sector clients feel it most because they operate in highly regulated environments and rightly so. Patient safety, public confidence, data privacy and responsible use of public money are the currencies they trade in.

‘Move fast and break things’ is not a viable strategy in a hospital. Processes must be proven safe and compliant before they can be adopted. Our role is to help our clients adopt new technology at pace without compromising on those fundamentals, putting the right methodologies, assurance frameworks and clinical governance in place so innovation and safety move forward together.

Another challenge our clients are grappling with is finding the right path to adopt AI and get real return on the investment. That came through clearly in PwC’s recent Global Performance Survey. The answer is moving beyond isolated proof-of-value pilots into AI-enabled process handling at scale, and examining growth opportunities rather than just cost savings. In healthcare, that translates to taking administrative burden off frontline and back-office staff, applying AI to the problems it’s genuinely better at. That’s where real value can be unlocked.

Are you noticing new and emerging trends so far in 2026?

AI is the story of 2026. It has moved out of novelty and experimentation into mainstream adoption. A significant share of the population now uses it daily, including clinicians and patients. People have seen what AI does for the customer service they get from their insurance company, their utility provider, or their streaming service and they expect the same standard from their healthcare provider. Chat tools mean patients arrive at appointments better informed about their own conditions, which is changing the dynamic of consultations. We’re also seeing native AI companies turn their focus to health, which will accelerate the pipeline of AI-driven products coming to market.

Have you any advice for professionals interested in a career similar to yours?

Stay positive. There’s a lot of pessimism in the headlines about disruption and changing business models, but disruption is often the biggest accelerator of opportunity. If you look beyond the displacement headlines, there’s a bigger story about the opportunities being created by AI.

This is the fifth major technology disruption cycle of my career, starting with the dot-com crash in 2001. I was an intern expecting a graduate job that didn’t materialise because of the crash. That led to a different job and a career that took me from Ireland to the US, Canada, India, the UK and a few places in between. 

Every cycle since, from the financial crash of 2008 to the cloud and SAAS revolution of 2011, even Covid, has been initially destructive but has ultimately created more opportunities for people in this industry. This is the reality we’re seeing as right now we are growing and hiring for our advisory health team.

My second piece of advice, don’t wait for doors to be opened for you. Ask. Tell people what you’re interested in, ask for new roles, ask for introductions, find out what’s possible. You manage your own career. That’s as true now as it has ever been.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

London Tech Week opened the way these events increasingly do: with a leaderboard of investment pledges. By the end of the first morning, the UK had collected several billion pounds in AI commitments, most of it aimed at the unglamorous machinery of compute.

Prime Minister Keir Starmer kicked off the keynotes with a new national AI compute strategy, including £400mn to buy specialist AI chips and expand the country’s computing capacity, part of a push he framed around keeping British firms able to “start here, scale here and stay here”.

The bigger numbers came from industry.

AMD committed up to £2bn over five years, backing high-performance computing with the University of Cambridge and Imperial College and taking direct stakes in UK startups, with chief executive Lisa Su on stage to announce it.

Cloud provider Nebius pledged around £1.7bn to build out UK AI capacity, funding three new deployments of Nvidia infrastructure that will reach 65 megawatts by 2027 and expanding its London R&D hub.

London’s mayor, Sadiq Khan, added a smaller but pointedly different commitment: £12mn to help the city’s small businesses actually adopt AI, through readiness checks and mentoring, rather than build it. The Prince of Wales is due to appear later in the week, tying technology to his Homewards anti-homelessness initiative.

The backdrop is a UK tech sector that, by Tech Nation’s count, is now worth £1.2tn, with British AI startups raising more than £8.2bn in venture capital in the first half of 2026 alone, close to half of all European tech investment by the prime minister’s reckoning. Europe’s IT spending is forecast to grow 8.2 per cent this year to $1.3tn, its fastest in half a decade.

For a country anxious about being squeezed between the US and China, the figures are a useful retort.

There is a familiar tension under the optimism. Much of the money is for compute infrastructure, and most of that infrastructure runs on American technology: AMD’s chips, and Nvidia’s hardware inside Nebius’s data centres.

The UK’s sovereign-AI ambitions, real as they are, still lean heavily on US suppliers, a dependence the same week’s launch of Cosine’s home-grown “Lumen Sovereign” model was explicitly designed to chip away at. Building capacity in Britain is not the same as owning the stack.

Still, for one morning at Olympia, the direction of travel was clear, and loud.

Between government money, a US chipmaker’s billions, and a cloud firm’s data centres, the UK is betting it can be the place where Europe’s AI gets built. The harder question, as ever, is whether that turns into companies, jobs, and breakthroughs that stay, or simply more rented compute.

London Tech Week runs until 10 June, with the pitches, and the pledges, set to keep coming.

security

Miasma worm shapeshifts, but cloud secret-scouting remains the goal

Microsoft’s GitHub has disabled over 70 repositories after they were reportedly compromised by a worm in the latest open source supply chain attack.

The code shack took down 73 repos within the space of 105 seconds after its alarms were tripped on Friday, June 5, after detecting signs of the Miasma worm infecting its projects, according to StepSecurity’s co-founder and CTO, Ashish Kurmi.

Users reported issues quickly on Friday, after visits to those repos all resulted in the same message displayed, indicating that they had been disabled due to terms of service violations. 

According to StepSecurity’s analysis, the attack kicked off after a compromised contributor account pushed a malicious commit to Azure/durabletask. The commit dropped configuration files that triggered remote code execution on machines when a developer opened the repo in an IDE or AI coding tool, such as Claude Code, Gemini CLI, and Cursor.

Several developers soon reported broken CI/CD pipelines, a support thread showed, although a moderator said at the time this was due to “an internal management issue.”

“The repo that most immediately caused issues was Azure/functions-action,” Kurmi wrote, used to deploy code to Azure. With it being taken down, every workflow that referenced Azure/functions-action@v1 stopped resolving.

GitHub stepped in a few hours after the repos were infected by the malicious commit. Its automated detections kicked in and disabled the repos in under two minutes, in two separate waves.

However, it was the borking of the durabletask family that hinted at the bigger picture, that the attack was indeed a re-opening of the previous Miasma worm attack that hit Microsoft last month.

Microsoft’s durabletask PyPi package was a previous target of the Miasma worm on May 19. Within a 35-minute window, three versions of the package were uploaded to PyPi, which planted infostealers on developers’ machines, specifically sniffing out cloud secrets and developer tool configurations on Linux systems.

Crucially, the re-targeting of durabletask suggests the tokens associated with the compromised developer account used to execute the PyPi attack were not fully rotated, allowing an attacker to gain access and push commits to GitHub, Kurmi said.

It was either that, or the contributor was re-compromised through the worm’s own propagation loop, or a different contributor’s token was used but the attacker altered the metadata to make it look like a repeated attack.

Security shop Snyk described Miasma as a descendant of the Mini Shai Hulud worm. It’s the same one that ravaged open source packages over at the npm registry, including Red Hat’s, earlier this month.

Cybercrime group TeamPCP claimed responsibility for developing Mini Shai Hulud, which itself is named after an earlier worm of the same name, sans “mini.” 

However, because TeamPCP open-sourced Mini Shai Hulud, it’s difficult to tell whether it was also behind Miasma or if someone else took the reins on the follow-up project.

StepSecurity also reported that two days before the Microsoft attack, the same worm was making a nuisance of itself at npm, compromising more than 50 packages, including a Vapi.ai SDK with more than 408,000 monthly downloads.

The Register asked Microsoft for comment, but it did not immediately respond. ®

Photo credit: Sonny Dickson
New photos of a production-style dummy device for Apple’s first foldable iPhone have surfaced online. They come from leaker Sonny Dickson and offer the clearest view of the design yet. Sonny Dickson, the leaker, provided images from several perspectives. The unit appears in both closed and fully open variants, with a screen installed for a more realistic appearance than the previous crude iteration.

When you turn the device off, it shows a compact exterior display that is about 5.5 inches across. The front-facing camera is located in the center, with a circular cutout at the top. If you are familiar with the traditional iPhone, you will see that this one is somewhat shorter and wider, similar to a passport held upright. However, when you fold it open, the experience changes dramatically. The internal display stretches to about 7.8 inches and has a broader, shorter aspect ratio, akin to an iPad mini. That leaves plenty of room for apps, videos, and split-screen work, assuming the software can adjust to the new configuration.

Sale

Google Pixel 10 Pro Fold – Unlocked Android Smartphone – Gemini AI Assistant – Advanced Triple Rear…

• Unfold extraordinary with Google Pixel 10 Pro Fold; with Pixel’s largest screen and Gemini, Google’s most advanced AI, it’s made for…
• Unlocked Android phone gives you the flexibility to change carriers and choose your own data plan[2]; it works with Google Fi, Verizon, T-Mobile…
• The gearless, high-strength hinge makes it durable enough to handle about 10 years of folding[3]; plus, Pixel 10 Pro Fold is built with Corning…

In the close-up images, you can notice a little round camera hole in the upper-left corner of the huge inner screen. Apple used a similar technique to previous Android foldables here, marking the first time this format has been used on an iPhone display. The vertical wrinkle down the middle of the interior screen is one of the most noticeable features. It appears faint on the dummy, but speculations suggest that Apple was able to reduce the depth and angle of the fold by changing the hinge. With some fancy glass and adhesives, the crease should be even harder to see on the finished units.

When the device is closed, a horizontal camera bar occupies a portion of the rear. Two lenses are side by side on a raised plateau that resembles the camera section of the iPhone Air, but no telephoto lens is seen at this time, and there is a small cluster of holes for the rear microphone. Other notable elements in the photographs include a titanium frame, volume controls relocated to the top edge, and a side button with Touch ID for unlocking and making payments, as this design does not appear to have Face ID.

Rumors suggest that there will just be one color option available at launch. White is the only color indicated in recent leaks, though we may see different colors later, as Apple has done with previous iPhone releases. It all comes together to create a device that fits in your pocket for everyday use and transforms into something resembling a small tablet when opened. The outer screen is ideal for fast peeks and calls without having to unfold the entire device, while the inner screen is far more comfortable for long-term tasks.
[Source]

Apple is holding a lot of secrets this year, and it’s about to share some of them with us. But what will today’s surprises bring?

The keynote address at the company’s Worldwide Developer Conference starts at 10 a.m. PT today at Apple Park in Cupertino, California. We’ll get a glimpse of Apple’s upcoming features and software updates, like previews of iOS and WatchOS, but likely won’t see any new hardware, like Apple’s rumored foldable iPhone, yet.

For now, WWDC 2026 is all about Apple’s software features and AI developments. And everything is still a big mystery. But that makes it fun to speculate about what might be coming. 

To lean into the excitement, the CNET Group is hosting a Big Guessing Game contest across its websites — CNET, Lifehacker, Mashable, PCMag and ZDNET. Three rounds of guessing let you predict what Apple will unveil this year. Each answer you get right earns you one chance in a drawing to win the latest Apple Watch announced in September. 

The Big Guessing Game Round 1 was all about Apple’s new software (like iOS, iPadOS, VisionOS, WatchOS and MacOS). Your guesses ahead of June’s WWDC event are in and are outstanding. 

Most readers think Apple’s new CEO will speak at WWDC

June’s WWDC keynote is expected to be the last one for Apple CEO Tim Cook before he resigns in September. Whether Apple’s incoming CEO, John Ternus, will speak at the event is still unknown. The vast majority of CNET readers (96%) believe Ternus will speak during Monday’s keynote, while 4% think he won’t. 

Here’s some food for thought. Cook was Apple’s chief operating officer before being named CEO. He was less involved with product development and announcements and instead created and refined Apple’s robust supply chain ecosystem, which helped Apple become a trillion-dollar company. 

Ternus is currently Apple’s vice president of hardware engineering, and has been a part of Apple event keynotes before. We’ll have to see if Ternus lands onstage at any point as a presenter, but it’s almost certain that Cook will kick off the event with his usual, low-key “good morning.”

MacOS 27 may not have a nickname this year

CNET readers didn’t come to a consensus on what MacOS 27 will be named, but there are a few guesses that stand out. Since 2013, MacOS has always had a nicknamed based on a California landmark, like Mavericks or Catalina. Prior to that, Apple’s Mac software had big cat names like Cheetah, Jaguar and Puma.

The names readers guessed for the new OS varied, but slightly over 9% believe it could simply be named MacOS 27 without a California nickname this year, which would be a first. Some readers think it could be Redwood (6%), Shasta (4%) and Mammoth (3%). Other names that stand out include Big Bear and Emerald Bay. 

Here’s a look at earlier versions. Keep in mind that Apple changed its numbering conventions in 2025 to reflect the year following its release. 

• MacOS 26 (2025): Tahoe
• MacOS 15 (2024): Sequoia 
• MacOS 14 (2023): Sonoma 
• MacOS 13 (2022): Ventura
• MacOS 12 (2021): Monterey 
• MacOS 11 (2020): Big Sur 

The frequency of ‘Apple Intelligence’ mentions is a toss-up

Based on guesses from CNET Group readers, the term “Apple Intelligence” could be mentioned anywhere from zero to a billion times during the WWDC keynote on Monday. Even among the reasonable guesses, there was a wide variation.

The median guess, or the middle number of all the guesses, was 26, but the average guess was much higher at 68. (There were some very high guesses.) 

We first heard the term Apple Intelligence at WWDC in June 2024, and it began rolling out to iPhones, iPads and Macs in October 2024. Since then, Apple Intelligence has become a buzzword for Apple amid the recent evolution of AI. 

Apple Intelligence launched with a few features, including text rewriting, photo cleanup to remove unwanted objects, and notification prioritization. However, AI has advanced significantly since 2024, and Apple is increasingly investing in its evolution. 

Recently Apple Intelligence updates to accessibility features were unveiled for hardware: VoiceOver and Magnifier, to describe what’s on your screen. Apple Intelligence now supports natural language, letting you speak like a regular person to control your iPad or iPhone. 

As Apple evolves to embrace AI, we’re likely to hear a lot about Apple Intelligence as it weaves into more OS features, like the overdue Siri update expected to run on Google’s foundational models for Gemini AI. We’ll need to wait to see exactly how much AI dominates Monday’s WWDC keynote.

Watch this: What to Expect From Apple at WWDC 2026 | Tech Today

02:49

Readers are certain WatchOS will get an AI health coach 

Speaking of AI, most CNET Group readers (89%) believe the next WatchOS upgrade could include an AI health coach, while 11% don’t believe so. WatchOS 26 came with a few fitness upgrades, notably Workout Buddy, which is an AI-powered personal trainer that acts in real time. 

Vanessa Hand Orellana, CNET’s lead smartwatch writer, said in her commentary that Workout Buddy could be Apple laying the groundwork for fitness and AI. It’s also a chance for Apple to see what it can do with AI on the smartwatch. Orellana added that an AI-powered coach helps you understand your health data and motivates you to build better habits. So a health coach that focuses on your overall well-being isn’t far-fetched based on what we’ve seen from Apple. 

Could the iOS 27 public beta drop as early as June?

I like that Apple gives us a chance to test out the new software before it’s officially released through its Apple Beta Software Program. But we never know for certain when the official public beta versions will be publicly available until they are. 

A surprising number of CNET readers — 53% — think Apple will release the public beta version of iOS 27 this month, while only 17% predict it will come in July. Others think it could come in August (9%) or September (21%). 

Even though it was the third-most-popular guess, there is a strong argument for July — the public beta of iOS 26 came out on July 23, 2025, and the public beta of iOS 18 came out on July 15, 2024. We’ll have to see whether Apple follows the same pattern and whether we’ll get any hints on Monday. 

Mark your calendars for July 7 for the second round of CNET’s Big Guessing Game, where we’ll focus on Apple devices, like Apple’s beloved iPad and iPhone. Remember, every correct answer gets you closer to a chance at winning the latest Apple Watch in September. 

Britain’s banks, telecoms, and weapons-makers have a new shared anxiety: that the AI they increasingly run on is built, owned, and controlled in the United States. A startup barely three years old is betting they will pay to fix it.

Cosine, a UK frontier-AI lab, has assembled a coalition of blue-chip British institutions to co-design Lumen Sovereign, what it bills as Britain’s first sovereign frontier AI model.

The signatories read like a roll-call of the national economy: BT, HSBC, Lloyds, NatWest, BAE Systems, Babcock, LSEG, PwC, Thales UK, Leonardo UK, and Telefónica Tech, each signing a memorandum of understanding to help shape the model’s use cases, security requirements, and governance.

The reveal was timed for the opening of London Tech Week, where Prime Minister Keir Starmer set out a more interventionist AI strategy and announced roughly £400mn of new spending on specialist AI chips to expand the country’s computing capacity. Britain’s next AI champions, he said, should “start here, scale here and stay here.”

The 💜 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

The selling point is control. Lumen Sovereign will be trained entirely on Isambard-AI, the Nvidia-powered Bristol supercomputer that ranks among Europe’s most powerful, using compute awarded under the government’s £500mn Sovereign AI programme, which named Cosine in its first cohort in April.

Cosine says the model will carry no dependence on foreign infrastructure at any stage, and can be deployed inside a customer’s own systems, including air-gapped environments with no connection to outside networks.

That is the crux of the pitch.

For most companies, choosing an AI model is a procurement decision; for a defence contractor, a bank running anti-money-laundering checks, or an operator of critical infrastructure, it is a legal and security one first. Sending classified systems, AML alerts, or clinical data to a server in a US data centre is often simply not allowed.

“Enterprises are increasingly waking up to the risk of being wholly dependent on foreign providers,” said Cosine co-founder and chief executive Alistair Pullen, who argues that vendor lock-in brings “security risk, dependency risk, and cost escalation risks.”

Cosine is an unusual candidate for the job, and a credible one.

Founded in 2022 by Pullen, Yang Li, and Sam Stenner, the Y Combinator-backed lab has raised just $8mn from investors including Lakestar, yet its coding models have topped independent benchmarks against OpenAI, Anthropic, Mistral, and DeepSeek for two years running. It supports more than 38 programming languages, among them the COBOL, Fortran, and Ada that still run Britain’s defence systems and financial plumbing, and that most AI tools handle badly.

Lumen Sovereign, it says, will be built from scratch on proprietary datasets spanning more than 30 regulated workflows, rather than fine-tuned from an open-source model, with deployment targeted for the end of 2026.

The coalition partners are blunt about why they are in. “Cosine has offered us a path to a completely UK native and highly customisable AI stack,” said Peter Passaro, director of AI and data at Babcock, pointing to “the very complex defence environments we operate in.”

Priority uses span cybersecurity, KYC and AML investigations, legal document review, and healthcare administration, areas where UK AI adoption has stalled on exactly these security concerns.

Cosine’s effort is the sharp end of a wider British scramble for AI independence, one that keeps colliding with an awkward fact. The government has poured money into homegrown compute and chip supply, and struck sovereign-infrastructure deals with Nscale and Nvidia, even as flagship US projects such as OpenAI’s Stargate UK have stumbled.

Yet “sovereign” AI in Britain still tends to run on American chips, Isambard-AI among them.

For now, these are memoranda, not contracts, and Lumen Sovereign is still to be built. Training a frontier model from scratch to the assurance standards demanded by defence and finance, and doing it by the end of the year, is a formidable task, and Britain has announced sovereign-AI ambitions before.

What is different this time is the coalition: a row of institutions treating sovereign AI not as a slogan but as a procurement requirement.