News

It feels like everyone and their aunt is making AI / AR smart glasses nowadays, especially as someone who tests the best smart glasses around. But something caught my eye when reading a description of Brilliant Lab’s new Halo glasses – as with their long-term memory capabilities, they promise to remind you of details of conversations and objects you’ve seen “years or even decades later.”

In real-time, Brilliant Labs’ specs can apparently offer contextual information based on what it hears and sees, too. This style of assistive help in the moment and later on sounds like a more ongoing version of features like the Ray-Ban Meta glasses’ visual reminders, features that Meta and others have said they plan to make (or have already made) an optionally always-on tool.

Now, Brilliant Labs has said its agent Noa will serve as a sort of AI VPN. Like a VPN reroutes your data to keep your online activity more private, Noa promises to offer similar levels of privacy as it communicates with the AI model powering its cognitive abilities.

Other Halo highlights are its “world’s thinnest AI glasses” design, its built-in display that sits in your periphery like some other AI specs we’ve seen announced this year, and it will have a relatively affordable $299 asking price (around £225 / AU$465) when it launches in November.

But even as someone who loves my Ray-Ban smart glasses and can see the benefits of these Halo glasses, I’m worried these smart specs are a sign we're continuing to race towards the death of privacy.

Smart glasses wearables with cameras are already, admittedly, something of a privacy conundrum. I think the Meta Ray-Ban specs do it well – only letting you snap pictures or short videos (or livestream to a public Meta account on Facebook or Instagram), and have an obvious light shine while you do so.

But the next generation of utility wants to boast an always-on mentality – cameras that activate frequently, or microphones that capture every conversation you have.

This would be like the Bee wristband I saw at CES (which Amazon recently bought), which promises to help you remember what you talk about with detailed summaries.

You can instantly see the advantages of these features. An always-on camera could catch that you’re about to leave home without your keys, or remind you that your fridge is getting empty, and Bee highlighted to me that you could use it to help you remember ideas for gifts based on what people say, or recollect an important in-person work chat you might have.

However, possible pitfalls are close behind.

Privacy is the big one.

Not just your own, though you’re arguably consenting to AI intrusiveness by using these tools, but the privacy of people around you.

They’ll be recorded by always-on wearables whether they want to, or even know they are, or not.

Privacy makes up a big part of media law training and exams that qualified journalists (like me) must complete, and always-recording wearables could very easily enable people to break a lot of legal and ethical rules. I expect that without these people necessarily realising they’re doing something wrong.

Big tech has always had an ask for forgiveness mentality. Arguably, because time after time, punishments (assuming they are even punished) are usually vastly outweighed by the benefits they reaped by breaking the rules.

This has seemed especially true with privacy, as our data seems to get mishandled by a company every other month – in small, but also sometimes catastrophic ways.

I’m looking at you Tea.

We’ve also already seen examples of AI companies playing fast and loose with copyright, and I expect the rulelessness will only get worse in the AI space as governments across the globe seem less than keen to properly regulate AI so they don’t hamper their country’s efforts to win the digital arms race.

AI wearables capturing every moment of our lives (from multiple angles to boot) with video and audio are a catastrophe waiting to happen.

Yes, there are always promises of privacy, and optional toggles you can switch on to supposedly enhance your data protection. Still, for every good actor that keeps its privacy promises, we can find plenty of companies that don’t – or quietly change them in new ToS you’re asked to sign.

We can hope that robust regulation and proper punishment for malpractice might come in and help avoid this disaster I foresee, but I’m not holding my breath.

Instead, I’m coming to terms with the demise of privacy – a concept already on its last legs – and accepting that while Big Brother might look different from how George Orwell pictured it, it will (as predicted) be watching us.

• Meta's next wearables announcement might include a smartwatch
• These AI smart glasses are giving me a feature I’ve been desperate for the Meta Ray-Bans to get since launch
• I adore my Meta Ray-Bans, but these new Oakley smart glasses are making me jealous

• Nvidia has announced that support for GTX 10 series GPUs ends in October 2025
• After that, these graphics cards, including the GTX 1060, will only get security fixes
• It also announced that Windows 10 support will run through to October 2026, mirroring Microsoft's extended support program for the OS

Nvidia has released a new graphics driver and announced that it'll soon be drawing the curtain on support for GeForce GTX 10 series GPUs, as well as GTX 900 models - and the end for Windows 10 gamers will follow a year later.

As Ars Technica highlighted, the release notes for driver version 580.88 came with the revelation that graphics cards based on Maxwell and Pascal architecture - meaning GTX 900 and 10 series products - will witness their final driver release in October 2025.

After that, they will only get quarterly security updates to patch them against vulnerabilities, and that's all. Security patches will finish in October 2028 for these products, too.

If October 2025 rings a bell, that's because it's also the month when Microsoft casts aside support for Windows 10, and that's also wrapped up in this Nvidia announcement.

Team Green said that it's extending Game Ready Driver support for Windows 10 to October 2026, to mirror the extended support Microsoft is offering consumers who want to stick with the OS, and not upgrade to Windows 11 yet. Or indeed people who may not be able to upgrade to the newer operating system, due to their PC not meeting the hardware requirements.

This move comes as no surprise, as Nvidia already told us back at the start of July that the v580 drivers would be the last to support Maxwell and Pascal graphics cards - we just didn't know exactly when the cut-off was coming, and now we do.

As stated, there are two categories of PC gamers who this affects: those with GTX 10 model GPUs, like the GTX 1060, and those running Windows 10. Further, some folks will be in both camps, no doubt - maybe quite a few.

GTX 10 series graphics cards are still reasonably popular in some cases (whereas GTX 900 products have pretty much dwindled away to nothing). In fact, the GTX 1060 is actually the 12th most popular GPU according to the latest Steam hardware survey - and once reigned supreme - so it's still seeing a lot of use.

After October 2025, this GPU, along with other 10 series offerings like the 1070 and 1080, will only receive security updates. That means they'll still be safe to use - patched against any exploits in drivers that may be found by the bad actors out there - but they won't get support for new games or features.

So, as time rolls on, you'll find that your trusty GTX 1060 becomes wonkier and less reliable with new games, as its final driver version ages and generally gets more erratic. Note that if you stick with old games, which were catered for before game support was frozen, you should be fine, at least in theory.

As for those on Windows 10, you'll be okay for another year yet. You'll still have full driver support through to October 2026, as noted, so you'll be fine until then. Assuming you keep Windows 10 itself secure, of course - using Microsoft's offer of extended support, which is now free, with a slight catch.

After October 2026, though, you'll need to upgrade to Windows 11, or you won't get new drivers – so no game support, or security patches either – no matter how new your Nvidia GPU is.

At this point, you're really looking at a Windows 11 upgrade - or a switch to something else entirely - unless Microsoft extends Windows 10 support further for consumers beyond 2026 (which seems unlikely, but could happen). In which case, Nvidia might again mirror the move with its own drivers - given that's what has happened here - but nothing's guaranteed by any means.

• Need a new GPU? Nvidia's RTX 5000 Super models may be coming sooner than you expect
• Nobody wants 8GB GPUs from Nvidia and AMD – and this retailer just made that clear
• AMD isn't giving up on high-end GPUs – a new leak hints at a new Radeon GPU challenging Nvidia's next-gen flagship graphics card

Agentic AI tools are being hired into businesses faster than we can fathom. They’re already helping shape workflows, influencing decisions, and generally assisting with the day-to-day running of a business. But without the right training, oversight, and integration, they risk becoming more of a liability to a business than a competitive advantage.

Unlike generative AI tools, agentic AI tools are designed to act independently, by identifying tasks, making decisions, and taking actions without direct instruction. They’re built to think and act, not just respond: becoming actors in the workplace, participating in workflows, integrating across functions, and shaping outcomes.

Most of the tools we see today are built for purpose or a specific outcome and are generally more like a bot that front-ends a workflow or business process. Today, we are largely at the task agentic worker level for automation, where we have repeatable tasks that are completed within very specific guardrails and there is very little gray area. Where this is gray area, there is a human in the loop that is making the ultimate decision or determination.

Traditionally, the role of a Chief Information Officer (CIO) is to manage and implement a company's information technology systems and strategies. As overseer of all IT operations, from hardware and software to data and IT infrastructure, the job is to ensure technology supports and drives business goals.

Today, managing agentic AI tools, CIOs are beginning to find themselves in less familiar territory. To manage these new digital entities responsibly and effectively, CIOs must step into a role that looks increasingly like that of an HR director, using onboarding and training, setting expectations for roles and responsibilities, and continuously evaluating their performance, just like you would with a new employee.

This might sound surprising, but if CIOs don’t adopt this mindset shift soon, they risk introducing not just inefficiency, but reputational and security risks across their business.

When a human employee joins a company, they're given training, clear expectations of their role, access to the right systems, and, crucially, guardrails. We should not be treating our new agentic AI co-workers any differently.

Many organizations believe that they can implement these tools quickly, with minimal oversight or structure - indeed many of the agentic AI vendors are marketing their technology with exactly these promises.

But agentic AI tools are dynamic: they evolve over time, learn from the data they interact with, and can behave in ways developers can’t fully anticipate. CIOs must therefore take the lead in designing structured, thoughtful onboarding processes for these tools, as well as ongoing reviews and feedback loops.

This includes defining their scope of responsibility, controlling what data and systems they can access, and outlining what constitutes successful performance. Without them, agentic AI tools can quickly create confusion, make flawed decisions, or even introduce bias or compliance risks.

When a human employee joins a company, they're given training, clear expectations of their role, access to the right systems, and, crucially, guardrails. We should be treating our new agentic AI co-workers exactly the same.

Like any other employee, agentic AI tools need clear success criteria, so CIOs should ask themselves: Is the AI doing what I expect it to do? And is it doing it well?

It’s critical that CIOs establish the right metrics to measure the performance of agentic AI tools. Primarily, CIOs must understand how agentic AI is improving workforce functionality by setting KPIs for tangible metrics, for example, speed and accuracy of execution, cost saved, and alignment with business goals. But performance should also be viewed through the lens of efficacy of outcomes: what are we willing to pay to achieve a specific outcome and what margin of error is acceptable?

This should not be a one-time thing. Annual performance reviews are inadequate for these new colleagues. Metrics need to be tracked and evaluated continuously because agentic AI can drift. As these tools learn and adapt, their behavior may change in subtle ways. Without regular check-ins, CIOs could find themselves managing a tool that no longer does the job well, or worse, one that has learned to manipulate its objectives or produce skewed results.

Ultimately, CIOs must understand not just how well an agentic AI tool is performing, but also whether it remains within regulatory boundaries. Similarly, just like humans, can that tool be corrupted to produce biased results? In many sectors, from finance to healthcare, a human must still make the final call. Agentic AI can assist, but cannot replace that responsibility.

As CIOs assess and evaluate the performance of agentic AI tools, they should also consider mapping how they will expand their use across the business. As new employees are not expected to immediately take over every function of the business, you should start with small deployments of agentic AI and expand over time. Just like you coach and develop co-workers to take on new responsibilities, there should be a considered and phased testing and roll out of the appropriate tools for different jobs.

It's about building pathways for responsible expansion. Consider how the AI tool could take on increasingly complex tasks, under increasing scrutiny, with increasing access to sensitive data and systems, but only when it’s earned that trust through performance.

Similarly to how you might bring in a separate employee with a different skill set, you should also consider bringing in different tools based on different models that might be better suited to your desired task. As you search hard to find the right person to fit the team, so too you should find the right AI to support that function.

That level of careful progression doesn’t slow down innovation. It makes it sustainable.

With generative AI tools, concerns are largely centered around data leakage. But, with agentic AI, the more pressing issue is access. These tools rely on deep integration across systems in order to deliver value, but too much access creates unnecessary risk.

Just as they do for human employees, CIOs must apply a principle of least privilege to agentic AI, granting access only to the systems and data required for the task. Excessive user privileges contribute to more than half of security incidents for nearly a third (32%) of organizations.

Permissions are often given for a specific use case and never revoked, leading to unchecked access that can be exploited, or cause unintended consequences. Agentic AI adoption offers a clear opportunity for organizations to reevaluate legacy access structures and introduce zero trust access controls that are fit for purpose.

Ultimately, what we’re seeing is the evolution of CIO responsibilities. Where once the role was focused on infrastructure, systems, and uptime, it’s now becoming one of HR / IT hybridity, and managing intelligent, autonomous systems that function like staff.

One thing is clear: agentic AI is coming. And if anything, it will soon be as embedded in business operations as cloud services or productivity tools. Agentic AI tools offer extraordinary potential, but only if they’re managed with the same discipline, foresight, and empathy we apply to human teams.

Today, CIOs have a unique opportunity to shape how these tools are brought into the business, not just as technology, but as contributors to outcomes, and innovation.

We list the best HR outsourcing service.

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

At Future Publishing we rely on advertising to keep bringing you the content you love to read.

The majority of the content on TechRadar is created solely by our editorial team, but on occasion, we also work with external partners to create content we hope our readers will find interesting and useful.

In some cases, advertisers support us in producing content. This content is labelled so you can see who has funded it and how it was created. We use the label on the page to clarify the advertiser’s involvement in the content.

Articles that are labelled “Sponsor Content Created With…” are paid for and reviewed by a commercial partner.

They may be produced by the client or by staff employed by TechRadar. This is commercial content and so is subject to the Advertising Standards Authority regulations in the UK and Federal Trade Commission regulations in the US.

Articles that are labelled as "Sponsored By…" or similar labels are independent editorial articles, created by writers employed by TechRadar that have been funded through the support of a commercial partner.

When planning this content, the editorial team may find alignment with a funding partner on the topic and the headline of the article but the article is not subject to any client review in advance of its publish date.

This content abides by the Editors’ Code of Practice from the Independent Press Standards Organisation in the UK and Federal Trade Commission regulations in the US.

If the commercial partner receives a sponsored section within a larger editorial article, that section will have a clear “Sponsored” label.

Articles that are labelled as "Preferred Partner," mean a commercial partner is offering a preferential affiliate rate to Future in exchange for greater prominence on the page, such as by highlighting a particular deal for a product that our journalists recommend.

TechRadar will only write content featuring “Preferred Partners” when we feel the content or product is aligned with what our audience wants. It is not sent to the funding partner for approval.

This content abides by the Editors’ Code of Practice from the Independent Press Standards Organisation in the UK and Federal Trade Commission regulations in the US.

• CMA says UK cloud market is uncompetitive
• AWS and Microsoft account for 30-40% of the UK market each
• The two companies disagree with the CMA's findings

The UK's Competition and Markets Authority (CMA) has determined Britain's cloud market shows too many anticompetitive traits, with Microsoft and AWS each holding around 30-40% of the UK market in 2024 and hyperscaler concentration especially high in infrastructure-as-a-service.

At the same time, fewer than 1% of customers switch providers annually and multicloud usage is rare (particularly among SMEs with more limited budgets).

The CMA has blamed high egress fees, incompatible interfaces, latency and skills gaps for widespread vendor lock-in, which is ultimately weakening competition.

Behind the two hyperscalers, Google accounts for just 5-10% of the market, with others like IBM and Oracle having even smaller shares. Although AI capabilities are yet to change market dynamics drastically, existing positions are likely to be amplified, thus the CMA has stepped in to ensure competition remains healthy.

In its Final Decision ruling, the CMA took the biggest hits at Microsoft over its unfair licensing practices, which make it costlier to run Microsoft software on rival cloud providers.

A Microsoft spokesperson told TechRadar Pro: "The CMA Panel’s most recent publication misses the mark again, ignoring that the cloud market has never been so dynamic and competitive, with record investment, and rapid, AI-driven changes. Its recommendations fail to cover Google, one of the fastest-growing cloud market participants."

"Microsoft looks forward to working with the Digital Markets Unit toward an outcome that more accurately reflects the current competition in cloud that benefits UK customers," they continued.

"The action proposed by the Inquiry Group is unwarranted and undermines the substantial investment and innovation that have already benefited hundreds of thousands of UK businesses," an AWS spokesperson added.

On the flip side, Google supported the CMA's findings: "The conclusive finding that restrictive licensing harms cloud customers and competition is a watershed moment for the UK."

Elsewhere in the industry, the CMA has been criticized for not acting fast enough and addressing persistent issues like cloud credits, lock-in and procurement bias.

"We urge the CMA to use the powers at its disposal now to address these harms, rather than embark upon a new investigation that may not give customers relief for years to come," Coalition for Fair Software Licensing Executive Director Ryan Triplette shared.

Looking ahead, the CMA's next step is to designate Microsoft and AWS with strategic market status (SMS) under the Digital Markets, Competition and Consumers (DMCC) Act, allowing it to impose legally binding, targeted conduct requirements on the two giants.

"A significant driver of high cloud computing bills is the consolidation of the market into a handful of players. Until recently, these companies have been the only game in town, so they’ve been able to set the rules of the market, for example, including egress fees for switching, long lock-in periods, and more. In fact, Gartner has observed that most customers spend 10% to 15% of their cloud bill on egress charges," noted Akamai's John Bradshaw.

"UK businesses are under huge cost pressures. We need to make it easier for them to switch cloud computing providers and find pricing options that better fit their balance sheets."

• These are the best cloud hosting providers
• We've listed the best cloud storage and best cloud backup platforms
• Microsoft hits back against UK competition lawsuits, slams AWS and Google once again