Technology

Google is developing Project Jarvis, a new AI agent that browses the web for you, taking the phrase “Let me Google that for you” quite literally.

The rumor comes courtesy of The Information which reports “three people with direct knowledge of the product” expect the new computer-using AI could be demonstrated as early as December “alongside the release of its next flagship Gemini large language model, which would help power the product.”

Just imagine asking Google Chrome to search the web and book a holiday for you, allowing Jarvis AI to take over as an automated personal assistant to get tasks done - it’s wild, almost like the web browsing equivalent of the self-driving car.

Jarvis AI is very similar to Anthropic’s recently announced improvements coming to Claude AI which let the model take control of your computer to browse the web, launch applications, and even use your mouse and keyboard. Imagine that, but confined specifically to Google Chrome.

The Information’s report also claims Google’s AI agent can “respond to a person’s commands by capturing frequent screenshots of what’s on their computer screen, and interpreting the shots before taking actions like clicking on a button or typing into a text field.”

While we could see Jarvis AI by the end of the year, the insiders who spoke to The Information claim it could be released to just a small group of early testers before a major rollout. As it stands, “The agent currently operates relatively slowly because the model needs to think for a few seconds before taking each action.”

Are we ready for AI agents?

AI agents, like Google’s Jarvis, are systems that can complete tasks without human supervision, whether that’s controlling a computer or simply marking emails as read. Most of the major players in AI are trying to launch AI agents that can help single users and businesses facilitate mundane computer tasks, but at the current stage, most are still just experiments.

How do you feel about AI browsing the web for you? I suspect the general consensus is that we don’t quite trust AI models to complete tasks without the concern of hallucination. That said, if these rumors are correct we’ll see exactly what Google has planned in just a few months, and whether or not it can convince the average consumer that it’s time we let AI take control of our web searches.

You might also like...

• Gemini 2.0 could launch soon
• What is Google Gemini?
• Gemini Live is the best AI feature I've used

Microsoft has announced changes to its senior leadership team across the EMEA market as part of its commitment to supporting new and emerging technologies such as AI and automation.

The company announced Clare Barclay’s appointment as President of Enterprise & Industry for EMEA, a role that will see her retire from her current position as CEO of Microsoft UK, which she has held since October 2020.

In her place, fellow Microsoft worker Darren Hardman will step in as CEO of Microsoft UK on November 1.

Microsoft EMEA leadership shake-up

Previously serving as Microsoft UK’s chief executie, Barclay’s successes include setting up high-profile partnerships with companies such as the London Stock Exchange Group, Vodafone and Sainsbury’s, with a focus on artificial intelligence.

In her new role, Barclay has been tasked with driving growth and digital transformation across 120 EMEA markets, working closely with partners and startups to encourage the adoption of AI tools.

Hardman, who joined Microsoft in May 2023, will lead the company’s operations through its next phase of growth. Having served as the UK&I CEO for Amazon Web Services, among other roles, Microsoft credited him for his extensive work in the UK tech sector. Before moving to Microsoft, Hardman worked with the government’s Digital Skills Council.

“Clare and Darren both bring unique strengths and insights to their new roles," noted Ralph Haupter, President of Microsoft EMEA. "With this strategic leadership transition, we will bring the best of our industry and AI execution to our UK and EMEA customers.”

The company also used the announcement as an opportunity to reinforce its commitment to Britain, adding that Microsoft employs around 6,000 workers in the UK with offices in Paddington, Manchester and Edinburgh, and a headquarters in Reading.

More from TechRadar Pro

• These are the best AI writers
• Aim high and find your next role with the best job sites
• Microsoft and UK government sign multi-year tech agreement

No cursed tombs are safe as Netflix has renewed Tomb Raider: The Legend of Lara Croft for a second season – and I can't wait to see the adventurer fire her dual pistols once again as she takes on a man-eating marine beast in the upcoming series' next season.

The show's renewal comes just weeks after Tomb Raider: The Legend of Lara Croft premiered on the best streaming service on October 10, with Hayley Atwell (Captain America: The First Avenger, Mission Impossible 7) providing the voice of the legendary archaeologist in this latest videogame adaptation.

Based on the famous videogame franchise first released in 1996, Lara Croft has since seen many iterations. From Keeley Hawes in the original video game series, to Angelina Jolie in the 00s movie adaptations, now she's going on another animated adventure in Tomb Raider: The Legend of Lara Croft season 2.

Tomb Raider: The Legend of Lara Croft was one of four new videogame adaptations showcased during Netflix Geeked Week and has received mostly positive reviews upon its release with 73% on Rotten Tomatoes. While this doesn't make it one of the best Netflix shows, it's still a fighting fit TV adaptation that reimagines her legacy story.

What can we expect in Tomb Raider: The Legend of Lara Croft season 2?

The first season was an origin story of sorts; one taking place after the events of the 2018 game Shadow of the Tomb Raider and saw the intrepid explorer confronts her traumatic past while unraveling an ancient mystery.

Season 2 is set to develop her character from a novice to her journey of becoming the legend she's known as today. Per a Netflix press release, season 2's official logline reads: "When adventurer Lara Croft discovers a trail of stolen African Orisha masks, she joins forces with her best friend Sam to retrieve the precious artifacts. Lara’s thrilling new adventure takes her around the globe as she delves deeper into the hidden secrets of Orisha history, dodges the machinations of a dangerous and enigmatic billionaire who wants the masks for herself, while discovering these relics contain dark secrets and a power that defies logic. Power that may, in fact, be divine."

Showrunner Tasha Huo teased some exciting details about Croft's character in season 2 to Netflix Tudum: “So season 1, thematically, is about Lara embracing how her dad dealt with grief, which was isolating. When we meet Lara, she’s very isolated … As we get into season 2, we’re trying to build Lara’s team over the course of the show, so she goes from isolated hero, who only wants to do things on her own - a lone wolf - to realizing, ‘Actually, I have this really cool team behind me.’"

Huo also hinted at a shark encounter in Lara's future, so there's high hopes that some classic Tomb Raider action is set to go down in season 2.

You might also like

• 8 huge questions I need Arcane season 2 to answer before the highly-rated Netflix show ends
• Cobra Kai season 6 part 2: Netflix release date, trailer, confirmed cast, plot synopsis, and more
• PSA: Here are Netflix’s secret codes to see all its horror movies and shows

Get up to speed on the rapidly evolving world of AI with our roundup of the week's developments.

Shopping for tiny people is intimidating. Good news: We did it for you.

We’ve just caught a sighting of Nvidia’s RTX 5000 graphics cards for laptops, which is hopefully a hint that these next-gen GPUs are inbound to be revealed soon, as the rumor mill already believes.

The mobile Blackwell GPUs were spotted by leaker Harukaze5719 on X, posted up as part of the PCI Device ID database, likely by laptop makers (most, but not all, of the IDs discovered by the leaker pertain to mobile parts).

2980 / GB10229c0 / GB1022c18 / GB203M, GN22 / 5090 Mobile2c19 / GB203M, GN22 / 5080 Mobile2c2c / N22W-ES-A1 / @mooreslawisdead leaked the picture5070 Ti Mobile5070 Mobile5060 Mobile5050 Mobilehttps://t.co/PGWZsH221XOctober 27, 2024

As you can see in the above post, these are supposedly the RTX 5090 mobile GPU, plus the RTX 5080, RTX 5070 Ti and vanilla RTX 5070, RTX 5060 and RTX 5050.

Those are the models Nvidia has on the boil in theory, but obviously take this with a liberal quantity of salt – and we don’t get any spec info here, just the purported GPUs that could make up the Blackwell family for gaming laptops.

The rumor mill has shifted somewhat in recent times to float the suggestion that we’ll get these laptop GPUs launched at CES 2025, so we could see some of them very soon – alongside RTX 5000 desktop graphics cards (possibly the RTX 5090, 5080, and the RTX 5070 on the desktop front).

Analysis: Fingers crossed for that RTX 5080 powerhouse

As noted, this is more evidence to add to the growing number of hints that we will indeed see next-gen Blackwell GPUs for laptops in just over a couple of months. It comes on top of a recent eye-opening revelation that we might see an RTX 5080 that uses the GB203 chip (the same one in the flagship RTX 5090 mobile) – which is what this fresh rumor suggests – and is nicely powerful as a result. Indeed, the RTX 5080 might be a big step up from the RTX 4080 mobile, if this pans out (which is perhaps why Nvidia is doing this – as the RTX 4080 for laptops was widely criticized as being a weak offering when it emerged).

If this latest info is right, it would appear to suggest that the rumored RTX 5080 is indeed the vanilla version of the graphics card, and not some kind of RTX 5080 Ti (which might be one explanation for how powerful it supposedly is). So, this keeps hope alive that the high-end for Blackwell laptops is going to be more impressive this time around.

The only variant (Ti or Super) mentioned at all here is the RTX 5070 Ti, in fact – though that doesn’t rule out Nvidia producing such spins at a later date. Team Green probably doesn’t yet know itself what route it’ll take when it comes to diversifying the RTX 5000 range later on down the line.

Via VideoCardz

You might also like

• Nvidia’s RTX 5090 now rumored to have superfast clock speeds – as well as being super-slim – could this GPU be too good to be true?
• Top gaming PCs: great rigs for serious PC gaming
• Best PC games: must-play titles you don't want to miss

From Wednesday to The Witcher, Netflix's fantasy roster doesn't disappoint.

Apple Intelligence shows a lot of promise, but it feels like a first step in what could hint at larger changes for the iPhone.

Light bulb security cameras offer a monitoring alternative with super-simple installation: Here are our favorite models.

There are other really good bingeworthy series on this platform, but I can't stop talking about this one.

Enjoy upgraded convenience and efficiency with these top-rated Apple HomeKit devices.

If you want to learn more about this all-foam premium mattress from Lull, here's what to know.

Level up your giftee's TV-watching experience with a streaming device that makes it easy to find the content they want. From Roku to Apple TV, we put nearly every device and smart TV system to the test to find our favorite streaming devices to gift this holiday season.

I asked ChatGPT security questions and found that AI models think Tesla can tap into your home security system -- among other fake details.

2025 could be a big year for iPads, as it now sounds like Apple might have refreshes for three models planned, namely the iPad Pro, iPad Air, and the base iPad.

This is all according to Mark Gurman, who recently claimed that we’d see the iPad 11 and a new iPad Air in the spring (meaning sometime between late March and late June), alongside the iPhone SE 4, and in his latest newsletter for Bloomberg he said the same again.

This time though Gurman added that the iPad 11 will apparently have roughly the same design as the iPad 10.9 (2022), but that it will get a speed boost – one that we imagine will also allow it to support Apple Intelligence.

In any case, these two tablets will be landing before too long if Gurman is right, but if you’re holding out for a new iPad Pro then you might be waiting quite a lot longer, as in the same article Gurman said he doesn’t expect this to land until either late 2025 or early 2026.

When it does, it will probably come running a new M5 chipset, but Gurman doesn’t expect other major changes, since the design of the current model is only around six months old.

The iPad Pro (2024) is still quite new (Image credit: Future) Mostly small upgrades

So while 2025 might be a big year in terms of the number of new iPads we see, it doesn’t sound like these new models will be massive upgrades on the current ones.

As such, you can probably safely buy the iPad Pro (2024) right now, since it’s likely still quite a wait for the next one, and possibly with only a new chipset to show for it. It might be worth waiting for the next iPad Air, since that could arrive in the first half of 2025, but the iPad Air (2024) is still fairly new, so again, we might not see many upgrades.

The most worthwhile of these products could be the iPad 11, since unlike the latest Air and Pro models, the iPad 10.9 (2022) doesn’t support Apple Intelligence, so even if this upcoming model does just have a new chipset, it could open up a whole new world of features.

As ever we’d take these claims with a pinch of salt anyway, but Gurman has a good track record, so there’s every chance these release timings will pan out.

You might also like

• iPad Pro 13-inch (2024) review: an absolute powerhouse of a tablet
• What is Apple Intelligence? The new Apple AI coming to your iPhone
• Best iPad: top Apple tablets to consider right now

Last week, Apple dropped a seriously heavy hint that new Macs are indeed about to arrive – multiple Mac-related revelations are expected over the course of this week, likely consecutive launches over the next few days, starting today – and we’ve heard more about the spec of the purported M4-toting machines that are inbound.

Mark Gurman brings fresh rumors about the M4 Macs in the leaker’s latest newsletter for Bloomberg, and there’s exciting news about the entry-level MacBook Pro.

Firstly, we’re told that of the two M4 chips currently powering the iPad Pro (launched earlier this year), only the 10-core CPU (with 10-core GPU) will be in the base-level MacBook Pro models. So, the other CPU (a slightly weaker spin with a 9-core CPU, plus 10-core GPU) may be in the (also rumored) Mac mini or iMac, but not Apple’s new laptops (take all of this with seasoning, but Gurman seems fairly confident in his assertions at this point).

The even better news the leaker imparts is that “at least some low-end Macs” will likely come with 16GB of Unified Memory as minimum, and our guess – as previous churning from the rumor mill indicates – is that this means the MacBook Pro 14-inch will have 16GB with the entry-level model. We can expect low and high-end MacBook Pro 14-inch models, and a new MacBook Pro 16-inch, Gurman notes (as the leaker has previously said), including ‘higher-end’ M4 flavors (M4 Pro and presumably M4 Max).

Gurman also tells us that we’re getting an iMac 24-inch with vanilla M4, and two spins on a redesigned Mac mini (featuring a choice of the vanilla M4 and M4 Pro chips). As already rumored, it’s going to be an even more compact Mac mini, with two ports on the front, and with some versions there’ll be three connectors on the back (five ports in total).

The leaker further notes that the iMac 24-inch getting a relatively swift refresh to the M4 – compared to the gap between the M1 and M3 versions of the all-in-one – shows how important the new chip is to Apple. Mainly because the M4 is key in complementing Apple Intelligence, with its AI prowess being considerably peppier (the M4 sports 38 TOPS – a measurement of AI acceleration power – compared to 18 TOPS in the M3 chip).

(Image credit: Future) Analysis: M4 for all Macs as part of the big AI drive

All Macs are getting the M4 treatment because of this focus on AI and driving ahead with Apple Intelligence, and that does make sense. (Early in 2025, probably in Q1 Gurman believes, we’ll see a MacBook Air with M4, and Mac Studio and Mac Pro M4 models will purportedly follow later next year – maybe in Q2, then Q3 of 2025 respectively).

The M4 will clearly be a significant leap for AI processing power, and also for gaming, Gurman says, even though the gains over the M3 will be “noticeable but not stunning” on an overall level. In other words, an upgrade from the M3 may not be hugely compelling, but coming from the M1, the performance boost will be as significant as when Apple first switched to its M series CPUs from Intel machines, Gurman observes.

It all sounds quite promising then, especially if Apple can load up the MacBook Pro 14-inch entry-level with 16GB of RAM – though Gurman appears to pour cold water on the idea of all Macs having 16GB as a baseline for Unified Memory.

Apple Intelligence can run with 8GB of memory, the leaker notes (as seen with the iPad Pro), but given that the AI feature uses up a fair bit of RAM, that could leave a Mac lean for other apps – and it’s also lacking for future-proofing to boot – so surely 16GB at this point is a bare minimum for a Pro-targeted laptop? The question is: will this make the entry-level MacBook Pro 14-inch considerably more expensive? We hope not.

We might see new Macs as early as later today, and the most high-profile reveal is surely the MacBook Pro M4, so it’d make some sense to either start – or finish – with those laptops. In just a number of hours, then, we might know everything about the new MacBook Pros – or perhaps the new iMac 24-inch if Apple goes in reverse order of importance.

You might also like

• I used to laugh at the Mac Mini but today I bought one
• What is Apple Intelligence? The new Apple AI coming to your iPhone, iPad and Mac explained
• Apple just declared a bunch of Macs obsolete – these are the affected MacBook and iMac models

There’s little doubt that we’ll see a Samsung Galaxy Z Flip 7 and a Samsung Galaxy Z Fold 7 in 2025 – and any lingering doubts now appear to have been dispelled, as codenames for these phones have leaked.

GalaxyClub (via GSMArena) claims to have received information about these phones, stating that they’re in the works with codenames of B7 (for the Samsung Galaxy Z Flip 7), and Q7 (for the Samsung Galaxy Z Fold 7).

So far, so unsurprising; but alongside these, GalaxyClub claims there’s also a device with the codename Q7M.

So what could this be? The site doesn’t claim to know, other than suggesting that it could be some variation on the Galaxy Z Fold 7, but it puts forward two possibilities. One is that this is the rumored tri-fold Samsung foldable phone, although we’d think that would launch as a completely different model, rather than as a variant of the Galaxy Z Fold 7.

The other suggestion GalaxyClub makes is that this could be a special-edition model, like we’ve recently seen with the Samsung Galaxy Z Fold Special Edition, which is currently exclusive to South Korea. But that phone was codenamed Q6A, following on from the Q6-codenamed Samsung Galaxy Z Fold 6. So logically a successor would have the codename Q7A, not Q7M.

The Samsung Galaxy Z Fold Special Edition is a slimmer, slightly better Z Fold 6 (Image credit: Samsung) Affordable or ultra-premium?

So what else could it be? The other two possibilities we can think of are either a relatively low-cost Samsung Galaxy Z Fold 7, or perhaps an ultra-premium Samsung Galaxy Z Fold 7 Ultra.

Of those, the latter might be more likely, since Samsung recently said it has no plans to launch a cheap Galaxy Z Fold model. But really this could be anything – or nothing at all, if this leak turns out to be wrong.

We might get more of an idea in July or August, as that’s when the Samsung Galaxy Z Flip 7, and at least the standard model of the Samsung Galaxy Z Fold 7, are expected to land.

You might also like

• It looks very likely that the S25 series will exclusively use the Snapdragon 8 Elite
• Best Samsung phones: top Galaxy handsets ranked
• Samsung Galaxy S24 FE review: an AI-powered phone that's light on sacrifices

Several benchmark refinance rates trended upward this week, but rates are still well below last year's highs.

A few key mortgage rates climbed higher. Here's what experts say is next for the housing market this year.

In its first Critical National Infrastructure (CNI) designation in more than a decade, the UK government has now announced that data centers should be afforded the same mission-critical status as energy and water systems in a bid to safeguard them against potential threats. This important step aligns with broader European policy, acknowledging the essential role these facilities play in safeguarding data, powering cloud services, and driving technological innovation. 

As reliance on digital services continues to grow, particularly with the surge of AI tools and cloud-based applications, ensuring the security and resilience of data centers has become a top priority for governments and businesses alike. The UK's focus on boosting protections against cyberattacks and system failures highlights the increasing recognition of their role in maintaining societal and economic stability.

However, securing data centers alone is not enough. These facilities are part of a much larger networked ecosystem. Without robust interconnection and network infrastructure linking them together, the data they store and process cannot flow effectively across borders, industries, or platforms. As the world becomes more dependent on digital services, the ability to ensure uninterrupted, high-performance connectivity between data centers and networks is critical – not only for organizations or communities, but for entire economies.

Is protecting data centers alone enough?

We are currently seeing a massive boom in data center absorption and new construction projects, driven by recent innovations in AI. In the US, for example, our recently published study shows that current construction projects and plans for future projects will see megawatt (MW) capacity in the market grow by 160% in the next couple of years to around 28,000 MW, up from a little over 11,000 MW today. And that doesn’t include any new construction plans from colocation providers or hyperscalers that might be announced next month! We see similar booms all over the world, including the UK.

Data centers house everything from critical business data to the applications that power daily life. But as industries shift more toward AI, machine learning, and cloud services, it’s not the data centers alone but rather in combination with connectivity that are quickly becoming indispensable. This is because the seamless exchange of data is one of the most effective economic drivers in the modern world. The surge in demand for cloud computing, combined with the increasing reliance on AI-driven services, has led to exponential growth in the amount of data generated, stored, and exchanged. This makes the physical and digital security of data centers vital to the smooth functioning of modern society – a fact which the UK, along with its European counterparts, has now formally recognized.

There’s more to this picture, however. It’s not just about the data housed within these centers – it’s about how that data is moved, accessed, and utilized. The value of a data center providing services to enterprises is directly tied to its ability to connect with others through resilient, high-performance networks. Without effective interconnection, even the most advanced data centers struggle to deliver on their promise of real-time applications and low-latency services. In an interconnected global economy, the failure to prioritize connectivity and interconnection security – as well as data center security – threatens not just technological progress, but economic stability as well.

It's time to talk about interconnection

We’ve established the indispensable role of data centers in our IT infrastructure, but in reality their true value is locked until they are seamlessly interconnected. Interconnection platforms – or Internet Exchanges (IXs) – are hubs that encourage the aggregation of networks and cloud services, housed in multiple data centers, and allows them to connect and exchange data directly, bypassing the public Internet. 

Data center and carrier-neutral platforms are independent of any single provider, enabling organizations to choose the best data centers and the best network routes, without being tied to one specific vendor. They also enable many more options for creating a geographically redundant infrastructure to ensure the greatest resilience in the case of outages. This setup reduces latency, increases reliability, and prevents long-term vendor lock-in, giving businesses the flexibility to adapt and scale their digital connectivity while ensuring continuous, high-performance data flow across different networks.

In Europe, the framework supporting data-center-neutral and carrier-neutral interconnection has set a global precedent. By separating infrastructure from individual service providers, the European model not only boosts resilience but also fosters competition, driving down costs while improving connectivity performance. This neutrality is quickly becoming vital to sustaining the smooth, uninterrupted flow of data that aspirational economies now require.

The US is now following suit. The recently issued "Roadmap to Enhancing Internet Routing Security" is a clear recognition that the strength of a nation’s digital economy hinges not only on data centers, but on secure, reliable interconnection. Over the last decade, the US has adopted a model where data exchange is not dependent on any single provider or location. Instead, neutral Internet Exchanges and interconnection platforms are now able to offer improved resilience and cybersecurity of these data flows. This movement reflects a much broader understanding that vendor-neutral platforms enable more resilient connectivity by enabling organizations to reduce points of failure in their infrastructure and improving redundancy across the network.

Let’s go deep

One of the key drivers of this shift in the US is the nation’s growing concern around the economic and security risks posed by legacy, fragmented, vendor-dependent infrastructure. In a push to alleviate those concerns, more than 80% of US Internet Exchanges are now neutral, enabling multiple data centers and service providers to be woven together, ensuring that even if one path experiences issues, data can still move freely through alternative routes. 

The importance of this cannot be overstated. As AI-driven applications increasingly require real-time data processing and ultra-low latency, any disruptions in connectivity can severely impact business operations and broader economic activity. So even with the most high-performance colocation and hyperscale data centers in the world, progress and innovation will still be hampered without the use of neutral points of interconnection. The decentralized, redundant nature of neutral interconnection platforms creates a more secure and reliable environment for data exchange, bolstering not just national security but also the competitiveness of digital economies on the world stage.

Designating data centers as critical infrastructure is a positive first step by the UK government, akin to protecting vital buildings in a city. But just as a city relies on roads and intersections to function and keep things moving, digital infrastructure depends on networks and interconnection platforms to keep data flowing smoothly. To truly safeguard connectivity, governments must not only secure data centers, but ensure that the entire interconnection ecosystem, including networks and IXs, is robust, resilient, and well-guarded.

We've rated the best dedicated server hosting.

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