TechRadar News

• The Google Pixel 10 and Pixel 10 Pro could cost the same as their predecessors
• The Google Pixel 10 Pro XL might cost more
• The Google Pixel 10 Pro Fold is said to cost less, with future models reducing in price even further

It’s not unusual for prices of an upcoming phone to leak, but what is unusual is to hear about prices for devices that are several years away, and that’s exactly what’s happened today, with a massive Pixel series pricing leak. This leak also includes the Google Pixel 10 line though, so let’s start with that.

According to Android Headlines citing “a source familiar with Google’s plans”, the Google Pixel 10 itself will have an unchanged price, at least in the US, making it $799 again. If that holds true elsewhere it will be £799 / AU$1,349.

The Google Pixel 10 Pro will also apparently retain its predecessor’s price, meaning a starting price of $999 (and likely £999 / AU$1,699). You will however apparently have to pay more for the Pixel 10 Pro XL, with that said to be roughly $100 more than the $1,099 / £1,099 / AU$1,849 Pixel 9 Pro XL.

The best news relates to the Google Pixel 10 Pro Fold, which could actually be cheaper than its predecessor. Apparently, this will start at around $1,600, down from the $1,799 Pixel 9 Pro Fold. That might mean a price elsewhere of roughly £1,600 / AU$2,450.

The Google Pixel 9 Pro Fold (Image credit: Philip Berne / Future) 2026 and beyond

And here’s where we get to the more future-looking news, as the same source claims that the Pro Fold model will drop again in price in 2027, bringing it to around $1,500 (possibly around £1,500 / AU$2,300). There’s no word of a price change next year, so presumably that will stay at $1,600, and in 2028 it will apparently remain at $1,500.

The source also claims that Google will keep offering its current selection of handsets – a standard phone, a Pro model, a Pro XL, and a Pro Fold, as well as the A-series – through to at least the Pixel 13 series in 2028.

And speaking of the A-series, this will apparently remain at $499 (likely £499 / AU$849) through at least the Pixel 12a in 2028.

All that said, we’d take these claims with a pinch of salt. Any leaks that look this far ahead could easily be wrong, as even if this is Google’s plan currently, that could change.

The claim that Google will gradually reduce the price of the Pro Fold line is especially surprising, since we haven’t seen that with other foldables so far, and with tariffs and inflation, that could be hard to achieve. But then this might be exactly what’s needed to make foldable phones mainstream.

We should find out whether the first of these claims are accurate in August, as that’s probably when the Pixel 10 series will launch.

You might also like

• Detailed Google Pixel 10 camera specs have seemingly leaked, pointing to an extra lens and worse sensors
• Best Pixel phones: these are the Google phones to buy right now
• Google Pixel 9 review: a fantastic phone* (*added fees may apply)

Marvel Rivals Season 2 is just days away, meaning you'll be able to jump into the next season of Marvel Rivals, and try out the new Heroes soon. There's a ton of changes coming to the game in the next few months, including a shift in how Team Ups work, and an overhaul of the missions system. If things pan out, it should be easier than ever to earn new cosmetics, and work your way through the Battle Pass.

What's new in Marvel Rivals?

(Image credit: NetEase)

Marvel Rivals is currently primed and ready for a new season. Developer NetEase has released the full patch notes for the next update, which will add new Heroes, maps, and more. There's still time to work your way through the current Battle Pass, which features an awesome Wolverine skin in its final tier. For now, three strategist teams appear to be the meta, though I'm hoping Season 2 will disrupt this, giving teams more choice in how they play. We'll have to wait and see.

In my Marvel Rivals review, I praised the game's huge suite of Heroes, and the surprising amount of depth found when trying out each one. With Season 2, there's set to be two new Heroes, X-Men legend Emma Frost and the delightfully villainous Ultron. We know more about the former, with the latter arriving later on in the season. Since its launch, Marvel Rivals has changed quite a bit, adding smart changes to Competitive Mode while giving players something to work toward. It's consistent updates like this that make Marvel Rivals a worthy inclusion in our best PC Games of 2025 list.

Here's everything you need to know about the next season of Marvel Rivals, including when the current one ends, and what to expect from the new Heroes that are being added. As release day draws nearer, I'll update this page with all of the latest news and info.

When is the next season of Marvel Rivals?

Marvel Rivals Season 2 will arrive on April 11, 2025. This means that it'll launch at the following time according to your timezone:

• April 11 at 5AM PT: West Coast US
• April 11 at 8AM ET: East Coast US
• April 11 at 1PM BST: United Kingdom
• April 11 at 2PM CET: Central Europe
• April 11 at 9PM JST: Japan

The times above are based on the Tweet linked here, which gives an estimated downtime of 2-3 hours. Given that we now know when the game will go down for maintenance, I've approximated when it'll be back up. This could differ on the day, depending on how the maintenance period goes.

When does the current season of Marvel Rivals end?

(Image credit: NetEase)

Marvel Rivals Season 1.5 will go offline for a few hours before the launch of the second season. Keep an eye on the official Marvel Rivals website for any changes. For now, here are the maintenance times for the scheduled downtime preceding Season 2:

• April 11 at 2AM PDT: West Coast US
• April 11 at 5AM ET: East Coast US
• April 11 at 10AM BST: United Kingdom
• April 11 at 11AM CET: Central Europe
• April 11 at 6PM JST: Japan

Marvel Rivals Emma Frost release date

(Image credit: NetEase)

Marvel Rivals will add Emma Frost at the launch of the new season on April 11. She's a new Vanguard, operating on the front line of battle by utilizing telekinetic abilities and sheer brawling firepower. NetEase has fully detailed the character in the video below, which shows off each of her abilities.

It appears that she'll have a decent bit of range, as well as the ability to apply her diamond skin and move in for close-quarters combat. She has an ultimate that damages multiple enemies at once with telekinetic beams while pulling them toward her.

There are some interesting details to Emma Frost's kit that are worth diving into. Firstly, her dash and choke slam ability ends with a sweep kick which knocks enemies back. This can be used to throw an opponent off of an edge for a quick kill.

Emma Frost can also place diamond shields, using them to block Ultimates like Iron Man's. This works similarly to Groot's shield walls, in that there's a fair bit of range to where you can place them.

Emma Frost joins Hulk, Doctor Strange, Venom, Captain America, The Thing, Groot, and Magneto as a Vanguard. Primarily, she will be used to tank damage on the frontline, though she clearly has a lot of potential as a brawler, and some support elements in her shielding abilities.

Emma Frost joins Hulk, Doctor Strange, Venom, Captain America, The Thing, Groot, and Magneto as a Vanguard. Primarily, she will be used to tank damage on the frontline, though she clearly has a lot of potential as a brawler, and some support elements in her shielding abilities.

I'm a Venom main in Marvel Rivals, so I primarily tend to go with Vanguards over the other two classes, so I'll definitely be giving Emma Frost a go when she launches.

We'll have to wait and see how her Team Up with Magneto and Psylocke plays out, but having a 2-tank team with Emma Frost and Magneto could be a good place to start in terms of working out a new meta.

Marvel Rivals Ultron release date

(Image credit: NetEase)

Now for Ultron, who won't launch until Season 2.5. If we look at the release schedule for the previous season, I expect Ultron to arrive sometime in May.

We haven't seen much from this upcoming character just yet, only a brief tease in the Season 2 reveal trailer. It's clear that he's set to be the villain of the season but in terms of how he'll play? Well, that's yet to be seen.

Marvel Rivals Season 2 Team Up Changes

(Image credit: NetEase)

Marvel Rivals Season 2 will bring in a bunch of new Team Ups, while removing existing ones. The first new one sees Emma Frost as the anchor Hero, with an ability called Mental Projection. This gives Magneto and Psylocke illusory projections, making it more difficult for the enemy to hit them.

Next up is Arcane Order, which sees Doctor Strange sharing his magic abilities with Scarlet Witch, enhancing her attacks. Then, Stars Aligned allows Winter Soldier to leap towards allies at critical moments.

Here are the key points on the changes coming to Team Ups in Marvel Rivals Season 2:

• Mental Projection: Emma Frost, Magneto, Scarlet Witch (NEW)
• Arcane Order: Doctor Strange, Scarlet Witch (NEW)
• Stars Aligned: Captain America, Winter Soldier (NEW)
• Gamma Charge: Hulk, Namor, Iron-Man (Doctor Strange removed, Namor added)
• Ammo Overload: Rocket, Punisher (Winter Soldier removed)
• Dimensional Shortcut: Magik, Black Panther (Psylocke removed)
• Chilling Charisma: Luna Snow, Jeff the Land Shark (Namor removed)
• Voltaic Union: Thor, Storm, Captain America (Removed entirely)
• Metallic Chaos: Magneto, Scarlet Witch (Removed entirely)

Marvel Rivals Season 2 balancing changes

In the Developer Vision video you can view above, some general balance updates are outlined for the future of Marvel Rivals. We've collected the key points in the list below:

• Hero balancing (buffs): Black Widow, Mister Fantastic, Scarlet Witch, Jeff the Land Shark, Peni Parker, Moon Knight, and Rocket Raccoon
• Hero Balancing (nerfs): Doctor Strange, Adam Warlock, Loki
• New Hero schedule: One per month from Season 3 onwards
• Maps: Map rotation will be in effect from Season 2 onwards (ranked), Hellfire: Gala (new map)
• Modes: New limited-time modes will be added over time
• Competitive: Map rotation, Bans added to Gold III, each player drops 9 divisions at the start of Season 2, minimum level requirement now 15, One Above All and Eternity ranks only allow duo queuing, Increasing the rate of individual performance vs match outcome.

Marvel Rivals next season FAQ Who was the last Hero to be added to Marvel Rivals?

The Thing and Human Torch were added at the launch of Marvel Rivals Season 1.5, rounding out the Fantastic Four in-game. The Thing acts as a Vanguard, while The Human Torch is a Duelist. Both are great options for newcomers, with The Thing being the perfect starter tank to learn.

Who are the next characters coming to Marvel Rivals?

The next two characters coming to Marvel Rivals are Emma Frost and Ultron. Emma Frost has been fully revealed, and she'll be a Vanguard with telekinetic and diamond-skin abilities. Ultron has been shown in trailers, but we've not seen gameplay yet.

What season is Marvel Rivals in?

Marvel Rivals is currently in Season 1.5. This started on February 21, and runs through until April 11. We'll then be in Season 2.

You Might Also Like...

• Will Marvel Rivals be on Nintendo Switch 2?
• Best free games to play in 2025
• When to expect the next season of Fortnite

When OpenAI made headlines with its ChatGPT launch in 2022, it was clear that its evolution into the workforce was only a matter of time. According to Sam Altman, technology is advancing so rapidly that we can expect this shift within the next few months. But is replacing humans with AI employees a genius business move, or akin to hiring unpredictable interns?

While human employees come with additional costs such as training, sick leave, and salaries, AI workers bypass these expenses. Instead, they introduce their own unique challenges that can be just as disruptive for employers.

Holidays vs. System Updates

Virtual workers don’t need to rest in the same way as humans, and it’s often seen as the biggest benefit of AI workers. AI doesn’t sleep and can work at all hours, unlike human employees who have set hours and need regular breaks. However, while AI never has the need for out of office replies, it does require plenty of maintenance and updates to perform.

For virtual employees, system downtime is the AI equivalent of a holiday, while planned maintenance or unexpected outages can disrupt operations just like a human taking time off for appointments or illness. The promise of 24/7 availability isn't entirely true if there are issues in the infrastructure.

According to a recent report by Splunk, unplanned downtime costs the world's largest 2,000 companies a combined $400 billion per year, averaging $200 million per company. From lost revenue, regulatory fines, and hidden costs like slower time to market and worsened brand reputations, downtime can be a very costly challenge.

So, although AI doesn't need a time off to recharge, ensuring its performing as it should needs to be a top priority for any organization introducing AI employees. AI models need to be regularly updated to avoid deteriorating in accuracy, leading to outdated or incorrect outputs. The recent shift towards retrieval-augmented generation (RAG) is an attempt to combat this, by enabling Large Language Models (LLMs) to source data from verified external sources, but it’s not an easy fix and still requires close monitoring.

Sickness vs Hallucinations

An AI employee won’t call in sick, but they can suffer from bugs and hallucinations that can be equally as disruptive. An AI hallucination occurs when a model generates incorrect or nonsense outputs, and this happens more than you would think. For instance, in 2017, Microsoft's AI chatbot Tay had to be shut down after it began generating offensive tweets less than a day after its launch.

Just like human immunity, different AI models vary in their robustness against bugs. OpenAI, Anthropic, and Google are all racing to reduce model hallucinations, but even top of the range models can still be unreliable at times. For organizations with AI employees, the priority should be treating these issues but also working to detect any potential future issues before they cause harm. Ongoing governance is therefore essential, in the same way as managing employee performance to identify any challenges or skills shortages.

Monthly pay vs. Updates

It’s a common misconception that AI is a pay-once, use forever deal. The cost of deploying AI solutions isn't set by a wide-open market; it's influenced by major players like Sam Altman and Google’s CEO, Sundar Pichai. As AI models evolve, it’s likely that businesses may need to invest further into costly updates or building entirely new systems to keep up with competitors. This ongoing investment could be the same as providing pay raises or professional development for human employees.

For example, ChatGPT recently launched a business-tier solution, and companies are already integrating AI co-pilots into their workflows. These tools are often tied to evolving licensing fees, cloud computing costs, and retraining requirements which means the AI you implement today could be severely outdated in just two years, requiring additional investment to maintain the same level of efficiency.

Training vs. Data input

In the same way that human employees need training to improve and develop in their role, LLMs need vast amounts of data to learn from to enhance performance. The more complex a model is, the more data it requires. For example, some models may need at least 10 million labelled items. It’s not just about the volume of data; the quality and relevance of this data are also crucial, as AI can only be as good as the data it is trained on. Poor data input means poor data output.

Linking abstract knowledge to real-world examples, also known as “Grounding AI” is a way to improve the ability of AI models to produce better predictions and responses. To do this, many companies are turning to RAG as a way of enhancing AI accuracy, by allowing models to pull from verified external sources instead of relying purely on pre-trained data.

Just like human development, training AI needs to be done regularly and consistently to get the best results. Especially as regulations around AI tighten and change, it’s crucial for companies to remain compliant with data protection laws, bias mitigation requirements, and industry-specific standards. Especially for businesses in highly regulated industries such as finance, firms using AI must align with regulations like the EU AI Act, ensuring models are transparent and accountable. This means regular audits, governance, and retraining.

So, what’s the verdict?

The truth is, it's not about choosing whether AI or human employees are better for a business. Instead, the real power comes by harnessing both. Issues such as governance, training, hallucinations and updates can be just as time consuming and expensive for businesses as managing human talent. The value of AI is in enabling humans to take more time for the things that only they can do; deeper thinking, creativity and building authentic human connections – capabilities that AI simply cannot, no matter how much it continues to evolve and advance.

Granted, implementing AI tools into business processes will free up time and improve efficiency, but that doesn’t mean that humans don’t still have a place. When the efficiency and intelligence of AI join forces with the creativity, empathy, and strategic thinking of humans, the possibilities are limitless. Together, they can drive innovation, boost productivity, and propel businesses towards long-term sustainable success.

We list the best task management app.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

• Nutribullet and McLaren have teamed up for a set of three themed blenders
• Each blender features McLaren's signature gray and papaya livery colors
• The original versions of all three received high scores from our reviewers

Nutribullet, maker of some of the world's best blenders, has teamed up with the McLaren Formula 1 team for a set of F1-themed soup and smoothie-makers. The collaboration might sound like a belated April Fools joke, but it's very real, and the three race-ready blenders are available to buy now direct from Nutribullet.

According to Nutribullet, the three blenders are "infused with McLaren’s signature papaya aesthetic – blending speed, precision, and innovation from the racetrack to the kitchen."

The trio include the Nutribullet x McLaren F1 Team Portable – a new version of the Nutribullet Portable Blender we reviewed last year. Our tester Victoria Woolaston praised its light weight, which makes it ideal for carrying to the office or gym, and its very reasonable price tag. Its battery life was also impressive, and she found it lasted around two weeks between charges in typical use.

She did note that it struggled somewhat with particularly tough ingredients, and it was sometimes necessary to run two blending cycles to incorporate difficult vegetables into smoothies, but no more so than other single-serve blenders at this price point.

(Image credit: Nutribullet, McLaren) Pulp position

Also on the starting grid is the Nutribullet x McLaren F1 Team Pro 900: a papaya and graphite-hued version of the more powerful Nutribullet 900. Our reviewer Emily Peck found that the 900 produced consistently good results, and noted that it's a particularly good pick for small kitchens.

"There's only one setting, but it offers you 50% more power than the original Nutribullet, with the speedy blade delivering super-smooth results," she said.

Finally, there's the Nutribullet x McLaren F1 Team Ultra. The original Nutribullet Ultra earned four and a half out of five stars in our review, producing silky smoothies and nicely textured hummus with ease. It's also one of the quietest blenders we've tested to date.

There's been a trend in kitchen appliances recently for subtle, natural shades like buttery yellow and leafy green, but if that's not your style then one of these blenders could add a welcome shock of color to your kitchen. You could even use one to whip up a quick papaya and coconut smoothie for refreshment during this weekend's Bahrain Grand Prix.

You might also like

• The Ninja Blast Max is so powerful it can obliterate frozen strawberries and even make kale smoothies delicious
• Nutribullet's new blender makes everything from nut milk to soup with a single touch
• I've finally found a green smoothie that doesn't taste like eating raw veg

• PowerColor has launched a new Radeon RX 9070 XT Red Devil Spectral White GPU at $899.99
• Its Hellhound model is available at $849.99
• Best Buy features even older RDNA 3 GPUs at high prices above their MSRP

If you thought the GPU market would get any better after recent pricing issues, think again: unfortunately, there's yet another huge price hike on, this time on an AMD Radeon RX 9070 XT partner card.

As reported by VideoCardz, PowerColor has launched an overclocked Radeon RX 9070 XT Red Devil Spectral White GPU, with a price point at $899.99 - which is a ridiculous $300.99 more than the card's $599 MSRP. This is featured alongside PowerColor's Hellhound variant on Micro Center, priced at $849.99.

It's a similar move to another board partner of AMD's: I've previously highlighted the XFX Radeon RX 9070 XT on Best Buy, which was being sold for up to $939.99 at the time, and we're now seeing a similar pattern with PowerColor models. The same applies to older RDNA 3 GPUs, with prices above MSRP despite being more than two years old.

While there are currently tariffs on imported goods in the US - which could make prices even more absurd in the near future - these price increases were happening long before Trump kicked off his tariff war. From the moment Nvidia's RTX 5000 series launched, price hikes across multiple retailers came into effect.

It's also worth noting that AMD's Radeon RX 9070 series GPUs were originally being sold at retail price (at least during the launch), and now that's changed entirely within the span of just a few weeks. It can be argued that this issue stems from a lack of availability and a shortage of GPU dies (especially with such a high GPU demand from consumers, and the new hunger for silicon from the burgeoning AI industry), but this isn't the first time retailers and AIBs have raised prices to ridiculous amounts.

(Image credit: DarwelShots / Shutterstock) Is the GPU market on the verge of a major crash?

Since the 2020 pandemic, average prices for GPUs have soared immensely, with multiple PC gamers tempted to overspend - so this isn't exactly a new occurrence. This current state of the GPU market is quite literally the worst I've seen, as I've stated before, and with tariffs now involved, it could be the start of a graphics market crash.

That's a big could be, though, as there is clearly still a large number of consumers willing to buy these inflated GPUs, as most are sold out across most retailers. Regardless, given how matters are now, I can only see prices continuously rising and leaving more and more PC gamers unable to purchase new hardware.

The RTX 5080 is already expensive at its retail price ($999 / £939 / AU$2,019), and partner cards take this up to $1,500 (based on Best Buy listings). As great as the RTX 5080 and RTX 5090 are, I'd argue that there are very few gamers who would be willing to spend these sorts of figures just for an enhanced experience with better performance.

I'd hate to see it, but we could soon reach a point where consumers opt out of any GPU upgrades: with Nvidia and AMD's focus on AI now very clear, we could witness a slow and painful dwindle in priority for gaming GPUs for both Team Green and Team Red...

You may also like...

• AMD's powerful Ryzen AI Max processors are on their way to desktop PCs - and a side effect could kill off discrete GPUs for gaming
• Nvidia's new Game Ready Drivers appear to be so bad that game developers are warning gamers to stay away
• Hang on handheld fans, we could be seeing AMD's FSR 4 upscaling on new handhelds after all - if this new RDNA 4 leak is legitimate

• A new leak suggests Nvidia's RTX 5060 Ti and non-Ti variant will stick with the previous gen's 128-bit memory bus
• This is contrary to other RTX 5000 series GPUs using higher memory buses
• The rumored use of GDDR7 VRAM may be the biggest highlight over their predecessors

Nvidia isn't done with its RTX 5000 series GPU lineup yet, as its unannounced RTX 5060 Ti and RTX 5060 desktop cards are still on the horizon. However, a new leak has supposedly uncovered more of their specifications.

As highlighted by VideoCardz, Nvidia's RTX 5060 Ti and RTX 5060 will use a 128-bit memory bus, the same as their predecessor, according to a leaked shipping description. This is different from the RTX 5070's 192-bit memory bus, along with the rest of the Blackwell GPU lineup, which is important for improving VRAM performance in GPUs.

While this may seem disappointing at first glance, the leaked specifications suggest both GPUs will use GDDR7 memory: the RTX 5060 reportedly has an 8GB VRAM GDDR7 capacity, while its Ti counterpart will have both 16GB and 8GB GDDR7 options. As VideoCardz hints, this could come in handy as a potential boost with much higher speeds and bandwidth over their predecessors, which use GDDR6 instead.

It's worth taking this with a grain of salt: we know that the RTX 5060 will be unveiled eventually, but there's no confirmation yet on Team Green's end, so it's worth waiting for official specs to come to light.

Ultimately, these entry-level GPUs and their chances of success in the GPU market will likely depend on pricing. With availability, scalping, and inflation being Nvidia's main issues with its RTX 5000 series launch, the graphics giant may have a hard time convincing consumers to opt for its lineup if the GPUs prove difficult to obtain at retail price.

(Image credit: Nvidia, Shutterstock) Regardless of how good or bad the new GPUs are, pricing will determine their fate...

It's a sad reality that PC gamers are facing with the GPU market currently in disarray - and with recent tariffs mixed in with scalping and high demand, I doubt it’ll end any time soon.

Unfortunately, it means the RTX 5060 Ti and non-Ti GPUs will likely suffer the same fate (especially if they turn out to be great options for budget gamers). We've seen third-party cards from both Nvidia and AMD selling above MSRP at multiple retailers; if Team Green doesn't make a reference card for either of the upcoming GPUs, it could be much worse, as consumers will be left entirely at the mercy of retailers and Nvidia’s manufacturing partners.

PC gaming can already be an expensive hobby, and all recent speculation points towards this getting worse: besides PC hardware, we now have game prices to worry about too, thanks to Nintendo's absurd $80 game pricing (no, I won't shut up about this).

I really hope Nvidia can surprise consumers with these new cards, as it certainly needs a kickstart for good momentum. It's not just been inflation and pricing, but drivers and missing ROPs have haunted Team Green since the RTX 5090 and RTX 5080 launch. Let's just hope a new launch can get us excited again...

You may also like...

• Nvidia RTX 5050 and 5060 leaks suggest these mobile GPUs could be pepping up budget gaming laptops soon
• Keen to buy an RTX 5060 Ti? Benchmarks and more rumored sightings suggest you don’t have long to wait for Nvidia’s next GPU
• Latest Nvidia RTX 5060 Ti GPU rumor suggests good news and bad when it comes to price

The debate over the future of AI continues to rage, with battle lines being drawn between Big Tech’s domination of AI development and grassroots, community-driven innovation – a situation that will set the tone for how the industry develops in the years ahead.

In its recent history, AI has been increasingly dominated by a handful of tech giants, creating what many believe is a monoculture where just a few major players control access and progress.

However, new government policies and open source initiatives could help restore balance in the industry. The significance of open-source AI gained global attention in January during the ‘DeepSeek tech crash.’ While headlines focused on falling AI stock prices, the other big story was the immense potential of freely available AI models that can be used, modified and distributed without restrictions.

So, how might this play out, and what kind of AI ecosystem should the industry aim for to ensure the technology delivers for everyone?

Open source – a route to ethical AI innovation

Few would argue with the notion that open source and AI are natural partners. Both thrive on collaboration and shared innovation, allowing researchers and developers to collectively accelerate the pace of innovation while also ensuring broader access to emerging technologies.

Open source principles will be key to AI's future health, from collaborative improvements and knowledge pooling to transparency. Deeper integration on an industry-wide basis can create the conditions required for fairer, more ethical innovation. In doing so, the AI ecosystem can avoid problems associated with the vendor lock-in, limited transparency and restricted access to foundational models associated with uncompetitive technology markets.

On the other hand, if AI development remains concentrated within relatively few dominant players, users will undoubtedly face higher costs. They will be offered fewer opportunities to tailor solutions to their specific needs. This situation will only intensify as more AI is integrated into IT infrastructure.

That’s why finding a sustainable path for open source is crucial, especially as organizations today face massive internal and external pressure to become AI-powered. In an open source environment, AI tools will give organizations the power to deploy and manage workloads at the enterprise scale. In particular, they will allow businesses to select and integrate their preferred large language models (LLMs) while fully owning their data and infrastructure.

The rise of open source AI also means organizations can leverage the surge in new infrastructure software innovation to fuel and enable digital transformation. These emerging technologies create new levels of freedom and flexibility. As organizations will embrace these qualities, they drive further demand for reliable, secure and enterprise-ready solutions. In this context, AI/open source-led innovation becomes a virtuous circle.

The UK is the third-largest AI market in the world and the UK Government has recently shared the AI Opportunity Action Plan to boost economic growth and improve people’s everyday life.

The impact of open source across AI development and implementation is already impressive. A recent study by McKinsey found that open source AI is widely adopted by 63% of organizations overall and 72% in tech, particularly by organizations leveraging AI for competitive advantage (40% more likely to use it), with key benefits including lower costs (60%).

However, a democratic approach to AI , fostering open source-led innovation, will be essential for the success of a thriving AI ecosystem. The advancement of open technology with openness and choice for customers will be a corner-stone for creating resilient digital infrastructures

The role of regulation

Looking more closely at the growing role of open source AI models, however, progress is not simply a matter of accessibility or ideology; the industry also needs effective regulatory guardrails. While AI governance is still at a relatively early stage compared to other tech sectors, there are some clues as to the direction of travel. Last year, for example, the White House acknowledged the benefits of open source AI, such as fostering innovation and allowing a broader range of actors to examine and improve models.

To build on this sentiment, the ideal scenario would be to commit to AI regulations that ensure AI can be accessed, used, changed and shared transparently by anyone. Now more than ever, enterprises require sovereignty, control and the ability to customize AI implementations – all of which are best ensured through open source.

Additionally, community-governed development is the best way to reduce the possibility of flaws, drive innovation and leverage AI’s full potential. The Linux Foundation project AI & Data for example fosters collaboration and provides trusted AI & Data infrastructure, uniting developers, organizations, and users.

With open source AI, customers also have full visibility into the software code and development process, increasing trust and bolstering AI, which is still a nascent technology. Community-based open standards can also address AI safety concerns, not least because engaging the community is one of the best ways to improve security protocols and resilience.

Clearly, this is a highly complex and nuanced situation where competing interests and priorities will come under further scrutiny. For those interested in a future where AI innovation takes place in a fair and ethical environment, collaboration and open standards will be key.

Check out our list of the best small business software.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

I always enjoy a chance to mess with AI video generators. Even when they're terrible, they can be entertaining, and when they pull it off, they can be amazing. So, I was keen to play with Runway's new Gen-4 model.

The company boasted that the Gen-4 (and its smaller, faster sibling model, Gen-4 Turbo) can outperform the earlier Gen-3 model in quality and consistency. Gen-4 supposedly nails the idea that characters can and should look like themselves between scenes, along with more fluid motion and improved environmental physics.

It’s also supposed to be remarkably good at following directions. You give it a visual reference and some descriptive text, and it produces a video that resembles what you imagined. In fact, it sounded a lot like how OpenAI promotes its own AI video creator, Sora.

Though the videos Sora makes are usually gorgeous, they are also sometimes unreliable in quality. One scene might be perfect, and the next might have characters floating like ghosts or doors leading to nowhere.

Magic movie

Runway Gen-4 pitched itself as video magic, so I decided to test it with that in mind and see if I could make videos telling the story of a wizard. I cooked up a few ideas for a little fantasy trilogy starring a wandering wizard. I wanted the wizard to meet an elf princess and then chase her through magic portals. Then, when he encounters her again, she's disguised as a magical animal, and he transforms her back into a princess.

The goal wasn’t to create a blockbuster. I just wanted to see how far Gen-4 could stretch with minimal input. Not having any photos of real wizards, I took advantage of the newly upgraded ChatGPT image generator to create convincing still images. Sora may not be blowing up Hollywood, but I can't deny the quality of some of the pictures produced by ChatGPT. I made the first video, then used Runway's option to "fix" a seed so that the characters would look consistent in the videos. I pieced the three videos into a single film below with a short break between each.

AI Cinema

You can see it's not perfect. There are some odd object movements, and the consistent looks aren't perfect. Some background elements shimmered oddly, and I wouldn’t put these clips on a theater screen just yet. However, the actual movement, expression, and emotion of the characters felt surprisingly real.

And, I liked the iteration options, which didn't overwhelm me with too many manual options but also gave me enough control so that it felt like I was actively involved in the creation and not just pressing a button and praying for coherence.

Now, will it take down Sora and OpenAI's many professional filmmaker partners? No, certainly not right now. But I'd probably at least experiment with it if I were an amateur filmmaker who wanted a relatively cheap way to see what some of my ideas could look like. At least, before spending a ton of money on the people needed to actually make movies look and feel as powerful as my vision for a film.

And if I grew comfortable enough with it and good enough at using and manipulating the AI to get what I wanted from it every time, I might not even think about using Sora. You don't need to be a wizard to see that's the spell Runway is hoping to cast on its potential user base.

You might also like

• How AI can turn your home video into a Hollywood blockbuster
• This AI can turn your mundane video into a special effects spectacular
• Turn your selfie into an action star with this new AI image-to-video feature

Google surprised me when I first experimented with its NotebookLM AI tool. The AI-powered research assistant offers a relatively intuitive and focused setup to make custom guides on any topic you can name. It's now poised to become even more accessible with a mobile app soon.

The app should be exciting for anyone who works with large volumes of information or just has an idiosyncratic learning style. NotebookLM basically takes any text, including Google Docs, PDFs, and even YouTube videos that have a transcript, unifies it, and recontextualizes the result as a summary, study guide, or other presentation.

As good as NotebookLM's summarization and ability to synthesize notes are, its most memorable aspect is the Audio Overviews it can create. Audio Overviews basically turns all that information you uploaded into a podcast hosted by two AI characters. The synthetic conversation between the two AI hosts can actually sound like people who have read what you uploaded and want to discuss it.

At first, I thought it was a gimmick, but it's really just a reformatting of information like any other. If you’ve ever felt like you'd absorb details from books and other text better in a BBC panel discussion, it's a great option.

It's also the most obvious reason to anticipate the NotebookLM mobile app. It’s perfect for multitaskers, commuters, and anyone whose eyes are too tired to stare at another screen. I’ve used it to refresh myself on interview transcripts before writing articles and just as a fun way to explain a confusing topic. When the app hits smartphones, the ability to do this on the go may be huge for anyone who wants to absorb complex information but doesn’t always have the time or focus to read it.

Pocket NotebookLM

Of course, Audio Overview is just one part of what makes NotebookLM worth your attention. Unlike many AI assistants, which often operate in a vacuum or require you to explain everything from scratch, NotebookLM has context from the content you share specifically. Its abilities are many, from basic summaries and the ability to answer questions to creating timelines, brainstorming new ideas built on the information, and instant cross-references.

That's useful on a computer. As a mobile app, there's a lot more opportunity to exploit NotebookLM's tools. Sure, you'll have needed to upload the information, but if it's there, you can instantly pull up that quote you can't quite remember or fact about the person you're about to meet.

Of course, it’s not perfect. NotebookLM is only as smart as the information you give it. If your sources are a mess, it won’t magically fill in the blanks. It also can't replace real research. Sometimes, it misses nuance or answers too confidently when the source it pulls from is hedging. But the fact that it’s transparent about the answers helps compensate. And, at least it will always offer something to listen to on your next long commute.

You might also like

• I tried using AI to create the background music for a podcast, but I may stick to music libraries for now
• I used NoteBookLM to help with productivity - here’s 5 top tips to get the most from Google’s AI audio tool
• Google will turn those long documents into your next favorite podcast

• This new Bloomin8 E-Ink Canvas frame comes in several sizes and lasts for a year on a charge
• You could easily place one on your wall and have it rotate through shared images via an app upload or through an SD card
• These screens also fit into a popular IKEA frame

Digital photo frames are eye-catching pieces of technology for the home, but they come with challenges like power cables and the need for constant connection to the cloud.

What if you could place digital images throughout your home – on every wall – and none of them needed wires or full-time internet connectivity? That would be the new Bloomin8 E-Ink Canvas from Arpobot.

Unveiled in a Kickstarter program this week, the new full-color E-Ink displays are designed to fit inside some standard art and photography frames, including the RÖDALM series from IKEA.

(Image credit: Arpobot)

Based on E-Ink's Spectra6 technology, the Bloomin8 rechargeable displays support full color and 200ppi of resolution. More importantly, they can hold an image indefinitely, which means you only need power to change the display image. A Bloomin8 E-Ink Canvas can operate, according to Arpobot, for up to a year on charge.

It's a rare consumer use for the all-color Spectra6 panel, which E-Ink has traditionally pitched for commercial use (think advertising displays). But it's certainly not the only one.

Bloomin8 will compete with the PocketBook InkPoster we saw at CES 2025. It also uses E-Ink Spectra6 technology, and its accompanying app offers a vast library of images.

When we wrote about it in January, there was no mention of AI image generation. We also don't know about their ability to support IKEA frames.

(Image credit: Arpobot) Load and lock-in

Images can be sideloaded through the Bloomin8's integrated SD slot, which I know is a rather retro function. In the years before connected devices, many of the original photo frames relied on SD cards for their photo libraries. However, Bloomin8 also accepts image updates from its dedicated app on your smartphone (iOS or Android).

Even so, since it's not connected to the cloud and doesn't need the Internet for its photos, the Bloomin8 E-Ink Canvas does not draw power while displaying an image. That's the benefit of E-Ink; once the image is drawn, it stays visible on the display even without additional power.

Image 1 of 4

(Image credit: Arpobot)Image 2 of 4

(Image credit: Arpobot)Image 3 of 4

(Image credit: Arpobot)Image 4 of 4

(Image credit: Arpobot)

The matte images we've seen thus far look quite good, though it does take a few moments for the E-Ink to fully resolve and saturate the image. It's also worth remembering that E-Ink is a reflective technology and, like a painting or a photo, is only illuminated if you cast light on it. We don't yet know how well the imagery will hold up in all lighting situations.

If you don't have any art or photos you want to use or have something completely else in mind, the Bloomin8 app can generate AI imagery that you can upload to one of the canvases.

The Bloomin8 panels come in three sizes: 7.3 inches ($167), 133 inches ($349), and 28.5 inches ($999), which means you have many options for mixing up your living room photo collage wall. These are all Kickstarter prices, and retail will be more expensive.

You might also like

• The best digital photo frame 2025: for any budget
• Netgear Meural Canvas II review: a huge-scale digital ...
• This awesome E Ink poster frame looks like the future of wall art, if you can stomach the price tag

• Apple TV+ has a new limited-time deal that's an excellent savings
• Whether you're new or returning, there's a good chance you can get three months of streaming for just $2.99 a month
• The deal arrives shortly after Severance wrapped and before Your Friends & Neighbors premieres

While it’s not as good as the promotion Apple TV+ kicked off 2025 with, this one is likely as close as we’ll get. Better yet, it applies to both new and returning subscribers who meet the qualifications.

Right now, you can get three months of Apple TV+ for just $2.99 a month, totaling $8.97, to experience all the films and TV shows on the platform, such as Severance, Ted Lasso, and The Studio.

You’ll need to act pretty quickly, though, to score this deal. Apple’s making this latest promotion live today – April 8, 2025 – through the end of the day on April 24, 2025. You’re also not eligible for the deal if you’re eligible for three months of the service, meaning you recently bought an Apple device or your Apple TV+ is billed through a wireless carrier like AT&T, T-Mobile, or Verizon.

(Image credit: Apple TV+)

Yes, there are specific requirements, but if you meet them, you’ll score access to a growing library of content for a pretty great price. Basically, you’ll need to have just not been a subscriber to the service within the last month or so, reading between the lines.

With the Apple TV+ subscription, you get access to TV shows, movies, and documentaries, including all of the titles on our list of the best Apple TV+ shows to watch. Additionally, you’ll be able to stream select Major League Soccer games that are available for all users and Major League Baseball games at the end of the week, thanks to ‘Friday Night Baseball.’

Right now, we’re in the midst of The Studio, starring Seth Rogen and a laundry list of other celebrities. It’s a show that piqued my interest and that of my TechRadar colleagues. Additionally, Your Friends & Neighbors with Jon Hamm premieres on April 11. Still, you’ll also find other shows available with all episodes like The Morning Show, Silo, Mythic Quest, Ted Lasso, and countless others.

Apple TV+ is, you guessed it, easily accessible on Apple devices like the iPhone, iPad, Mac, and Apple TV 4K through the Apple TV app. Still, it’s also available on Android and countless streaming players and smart TVs.

So, if you haven’t subscribed to Apple TV+ in quite some time or want to see what the hype about Severance or Ted Lasso is about, it’s an excellent time to give one of the best streaming services a try. You can score the $2.99 a month for three months subscription here.

You might also like

• Apple TV’s new Android app just got two great updates, but it’s still missing a big feature
• Everything new on Apple TV+ in April 2025: The Studio, Your Friends and Neighbors, Government Cheese, and more
• Apple TV+ has discovered the Fountain of Youth and you can jump in this May when the movie starts streaming

• OPHANIM combines everyday tech with high-end scientific imaging capability
• Antimatter detection is now possible using repurposed smartphone camera tech
• Gigapixel device could reveal fine details of particle collisions instantly

At CERN’s Antimatter Factory, the AEgIS collaboration is exploring a novel approach to detecting one of nature’s rarest phenomena: antiproton annihilation.

eeNews reports the team, led by Professor Christoph Hugenschmidt from the Technical University of Munich (TUM), has developed an experimental detector using sensors originally designed for mobile phone cameras.

Instead of creating a new sensor system from scratch, the researchers repurposed sixty 64-megapixel mobile camera sensors to form a 3.84-gigapixel array called OPHANIM, short for Optical Photon and Antimatter Imager. This composite detector can observe where antiprotons collide with matter, annihilating in a flash of energy.

From mobile phones to antimatter research

“For AEgIS to work, we need a detector with incredibly high spatial resolution, and mobile camera sensors have pixels smaller than 1 micrometer,” explains Francesco Guatieri, Principal Investigator at TUM.

To adapt the sensors for scientific use, the team had to employ intensive micro-engineering to remove the camera sensors of layers meant for mobile phone electronics.

“We had to strip away the first layers of the sensors, which are made to deal with the advanced integrated electronics of mobile phones,” says Guatieri. This process enabled the sensors to directly capture the light patterns associated with annihilation events.

Despite their origins, the mobile sensors do not compromise performance. In fact, the new detector offers a 35-fold improvement in real-time resolution over earlier methods.

“Previously, photographic plates were the only option, but they lacked real-time capabilities,” Guatieri adds.

“Our solution, demonstrated for antiprotons and directly applicable to antihydrogen, combines photographic-plate-level resolution, real-time diagnostics, self-calibration and a good particle collection surface, all in one device.”

The OPHANIM detector allows researchers to observe annihilation events in real time with a resolution of about 0.6 micrometres, fine enough to distinguish between different particles created during the process.

"This is a game-changing technology for the observation of the tiny shifts due to gravity in an antihydrogen beam travelling horizontally, and it can also find broader applications in experiments where high position resolution is crucial, or to develop high-resolution trackers," says AEgIS spokesperson Dr. Ruggero Caravita.

"This extraordinary resolution enables us also to distinguish between different annihilation fragments, paving the way for new research on low-energy antiparticle annihilation in materials."

The implications of this work extend beyond antimatter research. OPHANIM’s ability to track particles with such precision could benefit a wide range of experiments, all while offering a low-cost model built from existing consumer technology.

You may also like

• New insights into the Sun's corona emerge
• A new AI tool wants to make serendipitous scientific discovery less human
• AI Marie Curie or Robo-Galileo might make the next big science discovery

• 8849 Tank 4 offers a large 11,600mAh battery for endurance
• Supports triple rear cameras including night vision
• 8849 Tank 4 supports 66W fast charging and 25W reverse charging

The rugged smartphone market is evolving, with many brands now integrating large batteries and even portable projectors into their devices.

Chinese brand 8849 is known for such devices, having previously released models like the 8499 Tank 2 Pro and Tank 3 Pro.

The company has now announced the 8849 Tank 4, a business smartphone that features a 100-lumen 720p DLP projector. It supports laser-assisted auto-focus up to 4 meters, micro-laser ranging, night vision camera, and an RGB camping light.

Projector phones are carving a niche

The 8849 Tank 4 sports a 6.78-inch AMOLED screen with a resolution of 1200 x 2650 and a 120Hz refresh rate. Measuring 174.3 x 85.4 x 23.9 mm and weighing 538g, it is only slightly smaller than the best rugged tablets on the market.

This phone is powered by a MediaTek Dimensity 7300 chip, paired with 12GB or 16GB of RAM and 256GB or 512GB of internal storage, expandable via TF card.

It houses an 11,600mAh battery, which is about half the capacity of the Tank 3 Pro, but still large for daily use. The device also supports 66W wired fast charging. While it lacks wireless charging, it compensates with 25W reverse charging via its Type-C 2.0 port.

For photography, the phone is equipped with a triple rear camera setup, including a 50MP Sony main sensor, a 64MP OmniVision night vision camera with infrared fill lights, and an 8MP 3x zoom lens.

Like other rugged smartphones, this device supports IP68 water and dust resistance. Connectivity includes Wi-Fi 6, dual SIM support, and compatibility with GSM to 5G NR networks, offering download speeds of up to 2.34Gbps.

Samsung once dabbled in the projector phone category with the Galaxy Beam released over a decade ago. However, the company ultimately exited the space due to high costs and limited adoption at the time.

Nevertheless, with new offerings like the 8849 Tank 4, as well as similar products like Oukitel WP100 Titan and Doogee V Max Play, projector phones may be poised for a comeback.

You may also like

• These are the best rugged laptops around today
• And these are the best rugged tablets you can buy right now
• Japanese MSG maker is ramping up production of key material in semiconductor packaging

• Ulefone Armor 30 Pro combines ruggedness with productivity
• A triple-camera setup in a rugged smartphone isn't that common
• 64MP camera and a 4W speaker reflect a shift in what rugged smartphones aim to offer

The best rugged smartphones often focus on strength and battery life, but in recent years, they’ve started to include features such as portable projectors and even dual displays.

The Ulefone Armor 30 Pro is a dual-screen rugged device featuring a 6.95-inch FHD+ main display with a 120Hz refresh rate. On the rear, it sports a 3.4-inch secondary screen with a 960x412 resolution, protected by Gorilla Glass.

While the displays can’t be operated independently, both run on Android, and the rear screen allows users to take selfies and video calls using the main camera system.

Big battery and a MediaTek processor

The device is powered by the MediaTek Dimensity 7050 chipset, paired with 16GB of RAM and 512GB of internal storage. It supports 5G connectivity and AI-powered features.

As with many rugged devices, it houses a large 12,800mAh battery that promises extended usage. It supports 66W fast charging for quick top-ups and offers 10W reverse charging to power smaller devices like smartwatches or earbuds.

One major surprise feature is the Ulefone Armor 30 Pro's triple rear camera setup - which is pretty uncommon in this category. It includes a 50MP main sensor, a 50MP ultra-wide shooter, and a 64MP infrared night vision camera supported by four IR LED blasters.

The Armor 30 Pro is built for more than just survival in harsh environments. It features pogo pin connectors for accessories like endoscopes and microscopes, along with a dedicated shutter button for quickphoto capture.

Another unusual inclusion is a waterproof 4W loudspeaker, embedded in the hexagonal camera bump, that's capable of reaching up to 118dB. It’s an unusual choice for a device in this category, suggesting a shift toward entertainment and daily use, making it feel closer to a business smartphone than a traditional field device.

While not a typical productivity smartphone, the dual-screen layout and advanced sensor suite elevate its potential as a rugged tablet replacement for those who need portability with function.

The Armor 30 Pro will be available from April 14, priced at $379.99 on Aliexpress, $549.99 on the official store, €499.99 on Amazon, 33,015₽ on Ozon, and $499.99 on Mercado Livre.

You may also like

• These are the best rugged laptops around today
• And these are the best rugged tablets you can buy right now
• Has x86 lost the data center battle? Arm claims victory

• UK-based Fractile is backed by NATO and wants to build faster and cheaper in-memory AI compute
• Nvidia's bruteforce GPU approach consumes too much power and is held back by memory
• Fractile's numbers focused on a cluster of H100 GPU comparison, not the mainstream H200

Nvidia sits comfortably at the top of the AI hardware food chain, dominating the market with its high-performance GPUs and CUDA software stack, which have quickly become the default tools for training and running large AI models - but that dominance comes at a cost - namely, a growing target on its back.

Hyperscalers like Amazon, Google, Microsoft and Meta are pouring resources into developing their own custom silicon in an effort to reduce their dependence on Nvidia’s chips and cut costs. At the same time, a wave of AI hardware startups is trying to capitalize on rising demand for specialized accelerators, hoping to offer more efficient or affordable alternatives and, ultimately, to displace Nvidia.

You may not have heard of UK-based Fractile yet, but the startup, which claims its revolutionary approach to computing can run the world’s largest language models 100x faster and at 1/10th the cost of existing systems, has some pretty noteworthy backers, including NATO and the former CEO of Intel, Pat Gelsinger.

Removing every bottleneck

“We are building the hardware that will remove every bottleneck to the fastest possible inference of the largest transformer networks," Fractile says.

"This means the biggest LLMs in the world running faster than you can read, and a universe of completely new capabilities and possibilities for how we work that will be unlocked by near-instant inference of models with superhuman intelligence.”

It’s worth pointing out, before you get too excited, that Fractile’s performance numbers are based on comparisons with clusters of Nvidia H100 GPUs using 8-bit quantization and TensorRT-LLM, running Llama 2 70B - not the newer H200 chips.

In a LinkedIn posting, Gelsinger, who recently joined VC firm Playground Global as a General Partner, wrote, “Inference of frontier AI models is bottlenecked by hardware. Even before test-time compute scaling, cost and latency were huge challenges for large scale LLM deployments... To achieve our aspirations for AI, we will need radically faster, cheaper and much lower power inference.”

“I’m pleased to share that I’ve recently invested in Fractile, a UK-founded AI hardware company who are pursuing a path that’s radical enough to offer such a leap," he then revealed.

"Their in-memory compute approach to inference acceleration jointly tackles the two bottlenecks to scaling inference, overcoming both the memory bottleneck that holds back today’s GPUs, while decimating power consumption, the single biggest physical constraint we face over the next decade in scaling up data center capacity. In fact, some of the ideas I was exploring in my graduate work at Stanford University will now come to mainstream AI computing!”

You might also like

• Bye Nvidia! Meta tests its first in-house training AI-PU
• TSMC, Broadcom could tear apart Intel's legendary business after 57 years
• Nvidia is dreaming of trillion-dollar data centres with millions of GPUs

• Two US Senators have asked tech giants for more information on AI deals
• Even partial acquisitions could harm competition
• The FTC also noted potential competition issues

Two tech industry titans that have been the subjects of many an anticompetitive investigation are now being questioned over their partnerships with AI companies.

Democratic US Senators Elizabeth Warren and Ron Wyden are reportedly demanding information about the deals Google and Microsoft have with companies at the forefront of artificial intelligence.

The concern is that these strong partnerships could impact competition within the industry, violate antitrust laws, and ultimately lead to both fewer choices and higher prices for customers.

Microsoft and Google questioned over AI deals

In question are Microsoft’s deal with OpenAI and Google’s deal with Anthropic, with the Senators seeking financial details of payments made by AI firms to their cloud providers and information on whether the companies have exclusive rights to certain AI models.

Warren and Wyden also want to know whether Google or Microsoft have any intentions to acquire their respective AI partners.

“Partnerships between CSPs and AI developers, if left unchecked, may accelerate consolidation of the AI sector, ultimately driving up prices and choking off innovation,” the two said in their letter to Google CEO Sundar Pichai and Anthropic CEO Dario Amodei (via Reuters), and in their second letter to Microsoft and OpenAI’s CEOs, Satya Nadella and Sam Altman.

This isn’t the first time these specific companies have come under fire for their involvement with leading AI firms.

A separate January 2025 report by the Federal Trade Commission had already studied such partnerships, raising concerns about potential acquisitions which could spell out disaster for competition.

“As noted in the FTC and Department of Justice (DOJ)’s merger guidelines, even partial acquisitions may present 'significant competitive concerns' because of the effects on firms’ incentives and strategy,” the Senators continue.

TechRadar Pro has contacted all four companies for comments and context, but we have not received any responses as yet.

Via Reuters

You might also like

• These are the best AI tools and best AI writers
• Check out our roundup of the best cloud hosting providers
• Microsoft says it found out about FTC probe by watching the news

• A critical flaw was discovered in file transfer tool CrushFTP
• Experts claim the issue was being abused in the wild
• CISA added the flaw to its KEV catalog

A critical-severity vulnerability plaguing file transfer software CrushFTP was found being actively exploited in the wild.

Earlier this month, it was reported that the software, commonly used by organizations to handle large-scale file transfers, contained an authentication bypass vulnerability which allowed unauthenticated attackers to gain administrative access.

By specifically targeting the crushadmin account, threat actors could abuse the flaw to compromise the target system entirely.

Monitor your credit score with TransUnion starting at $29.95/month

TransUnion is a credit monitoring service that helps you stay on top of your financial health. With real-time alerts, credit score tracking, and identity theft protection, it ensures you never miss important changes. You'll benefit from a customizable online interface with clear insights into your credit profile. Businesses also benefit from TransUnion’s advanced risk assessment tools.

Preferred partner (What does this mean?)View Deal

CISA adds it to KEV

The flaw is now tracked as CVE-2025-31161, and was given a severity score of 9.8/10 (critical)

It affects CrushFTP versions 10 before 10.8.4 and 11 before 11.3.1. Users are strongly advised to update to these versions immediately, and if they can’t, enabling the DMZ proxy instance can serve as a temporary workaround.

Security researchers have warned that the bugs were used in the wild to install remote management tools like AnyDesk and MeshAgent, The Hacker News reported.

CISA has also picked up on the news, adding the bug to its Known Exploited Vulnerabilities Catalog (KEV). This means that Federal Civilian Executive Branch (FCEB) agencies have a three-week deadline (until April 28) to apply the patch, or stop using CrushFTP entirely.

Cybercriminals often target managed file transfer software vulnerabilities, since they could allow access to sensitive corporate files and databases. In fact, one of the most devastating cyberattacks in recent history happened in 2023, when ransomware operator Cl0p abused a previously unknown SQL injection vulnerability in MOVEit managed file transfer software to breach hundreds of corporations around the world.

A year before that, GoAnywhere MFT was breached and used to steal sensitive data from almost 130 organizations, and in January 2024, the same software was found to be vulnerable to a critical path traversal weakness flaw.

You might also like

• Popular file transfer software has a seriously dangerous security bug that gives anyone free administrator rights — so patch it now to avoid another Moveit-like debacle
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

Hello and welcome to our live coverage of Google Cloud Next 2025!

We're live in Las Vegas and ready for a packed few days hearing about Google's latest cloud and AI news.

The event kicks off tomorrow (Wednesday) with a keynote hosted by Google Cloud CEO Thomas Kurian, who will be joined on stage by a host of guests, doubtless with much to tell us, so stay tuned for all the details as it happens!

Good morning from sunny Las Vegas! We're here for Google Cloud Next, and after a slighted delayed arrival into the city last night, are busy getting ready for the event kicking off tomorrow.

If you'd like a reminder of everything announced at last year's event, you can revisit our live blog here.

• Hades 2 will be a Nintendo Switch and Switch 2 exclusive at launch
• Supergiant Games announced the news alongside a new Creator's Voice video shared today
• Hades 2 will be coming to other platforms at a later time

Supergiant Games has announced that Hades 2 will be a Nintendo Switch and Nintendo Switch 2 exclusive at launch.

Hades 2 has been in Early Access for almost a year now and we recently learned that the sequel to the popular roguelike game will also be heading to the Nintendo Switch 2 when it's ready for a full release.

Now, alongside a new Creator's Voice video showcasing the behind-the-scenes development of the game, Supergiant has revealed that the game will be exclusive to the Switch and Switch 2 when it releases but that it's still planned for other consoles and PC post-launch.

Supergiant shared the news on X / Twitter, saying, "While we have no set date just yet for v1.0 of HADES II, we can confirm its console launch will be on #NintendoSwitch2 as well as the original Nintendo Switch, where HADES got its start."

Battle beyond the Underworld using dark sorcery to take on the Titan of Time in #HadesII. This bewitching sequel is coming first on console to both Nintendo Switch and #NintendoSwitch2. pic.twitter.com/dpsJZ56hLvApril 2, 2025

During the Nintendo Switch 2 Direct last week, a bunch of other Switch 2 exclusives were also revealed, including Mario Kart World, Donkey Kong Bananza, and The Duskbloods.

We also learned that the console will be released on June 5, 2025, for $449.99 / £395.99 or $499.99 / £429.99 for a Mario Kart World bundle.

Switch 2 pre-orders are officially live in the UK, but fans in the US will need to wait a little longer to secure theirs as Nintendo has delayed Switch 2 pre-orders from its original April 9 date to assess the potential impact of tariffs imposed by the Trump Administration.

You might also like...

• Here are the Nintendo Switch 2 launch games that you'll be able to pick up and play on June 5
• Nintendo confirms that certain Switch 2 game cards will just have a download key, but I don't think it's as bad as we first thought
• The Nintendo Switch 2 will feature DLSS and ray tracing, but we don't know which games support it

• Turkey plans to expand the authorities' power to block access to social media on the grounds of "national security"
• Global companies with over 1 million users in Turkey will also be required to set up a local company
• Messaging platforms are also included in the so-called "over-the-top service providers" expected to follow the new rules

Turkey plans to expand authorities' power to block access to social media and messaging platforms on "national security" grounds without a court order.

As per a report by the Turkish news site Ekonomim and reported by Bianet English, the draft amendment would also require these so-called "over-the-top service providers" to establish a local company when they have more than 1 million users in Turkey.

The draft law follows a 42-hour social media shutdown, amid protests sparked by the arrest of Ekrem İmamoğlu, that forced citizens to turn to the best VPN apps. Unrest is still ongoing across the country at the time of writing.

Turkey's internet control

Turkey is infamous for seeking control over the internet. VPN provider Surfshark, for instance, recorded a total of 17 instances in which authorities enforced some form of internet restriction since 2015. Instagram, X, YouTube, WhatsApp, and Telegram are the main targets.

Examples include a block on X at a time when people needed it most – the aftermath of the deadly earthquake that shook both Turkey and Syria in February 2023. Social media and messaging apps went dark during times of political turmoil in 2024, too.

Do you know?

(Image credit: Getty Images)

Turkish authorities have also been cracking down on VPN services, the most popular tool to bypass government-imposed internet restrictions, since at least December 2023. People have reportedly managed to evade this ban, however, as most VPNs, like Proton VPN, keep recording usage spikes across the country.

Halting the spread of hate speech and disinformation is usually the motive behind these temporary blocks to people's online communications, as per current regulations upon obtaining a court order.

The draft amendment would, however, empower the Information and Communication Technologies Authority (BTK) to directly restrict access to social media and messaging apps "on grounds of national security, public order, public health, or similar public interest considerations," Bianet English reported.

As the news site points out, BTK already has the power to throttle bandwidth by up to 90 percent for up to 24 hours during emergencies, de facto making accessing targeted websites and apps almost impossible.

The amendment could also expand on existing requirements to appoint a local representative, mandating a new obligation to set up a local company for providers with more than 1 million users in Turkey.

Mass censorship & information control is spreading worldwide. Turkey’s new “cybersecurity” law is widely seen as another censorship tool.Setting up a @signalapp proxy server helps keep private communication alive even when regimes try to shut it down.https://t.co/hpfgwQtrut pic.twitter.com/2Fl7doIEA0March 30, 2025

The news of the draft amendment comes at a delicate time of political turmoil. Protests over the arrest of the major opposition leaders are still ongoing, in fact, with journalists increasingly facing arrest.

X accounts of protesters and opposition figures have also been reportedly blocked at Erdogan's orders.

Additionally, on March 13, 2025, a new cybersecurity law was also enforced, introducing new powers to fight back against cybercrime in the country. These include "two to five years in prison for falsely claiming that a cybersecurity-related data leak has occurred," Bianet English reported.

Yet, digital rights now fear these new powers are too broad and could be used to restrict independent reporting and stifle dissent, the Global Voices reported. This could de facto increase internet censorship levels even further.

You might also like

• Is the future of censorship-resistant VPNs, no VPNs?
• 2024 was the worst year on record for internet freedoms – again
• Myanmar enforces new cybersecurity law – and VPN usage is the main target