Feed aggregator

• AirPods with cameras are apparently still coming...
• ... but the next model, due in 2026, won't have them
• In truth, little is known about these new buds (which is how Apple likes it)

For some reason, Apple fans are clamoring for new AirPods with cameras built in, and we've been hearing plenty of rumors that they're on the way. But a new report suggests they won't actually be the next Apple earbuds we see.

According to stalwart Apple correspondent Mark Gurman (via his Power On newsletter), Apple has two AirPods models on the go.

The flagship ice-white buds we've been reporting on are set to have tiny AI cameras — and Gurman has now backed up rumors of their being slated for a 2027 release.

But some models are expected in 2026: Gurman mentions another pair of buds (with a lot less known about them) that are likely to come along the 2026 iPhones at the end of the year.

The coverage doesn't specify whether these are AirPods, AirPods Pro or even over-ears, but the latter seem unlikely, because the AirPods Max 2 are only a few months old.

Could it be a possible Beats release? TechRadar's Jacob Krol recently spent a few days at Apple's Beats HQ and pressed the highly successful Apple subsidiary brand on some much teased Beats headphones, spotted on various occasions at the World Cup recently, but the company remained tight-lipped on those.

For now, we'll have to wait and see.

Are cameras worth the wait?

Questions concerning whether AirPods with cameras would even work aside, Apple faces the question — given the recent backlash over glasses that see — over whether most buyers would even want that.

The point of the tech is, apparently, to let an AI assistant see what's going on around you. The hardware is part of a big AI push Apple is undergoing.

That's despite consumers' appetite for AI dropping as fast as companies' desperation for AI increases. I don't see many AirPods buyers opting for a new pair of earbuds without a genuine reason they'd be useful (and no, smarter Siri doesn't count), especially if their whole schtick is that they're always spying on you.

Often when we see a tech product delayed, it's because it's just not ready yet. But with the camera-toting AirPods, it's more likely that we're not ready yet... or maybe we just don't need them.

• Microsoft has summarized its campaign to fix Windows 11 so far
• The company also let us know what work is coming, including making the OS run better with 8GB of RAM
• That should have been the case from the start, though – and there's a missing piece of the puzzle to address as highlighted by Windows 11 users

Microsoft has made it clear that it's working to ensure that Windows 11 runs better on 8GB of system memory, as well as letting us know about other elements of the OS the company is going to fix as 2026 rolls on.

This was part of a blog post which is an update on the progress made with the campaign to fix Windows 11 that kicked off in March, summarizing all the work Microsoft has done – and there's been a lot of that, without a doubt. (I recently picked out the most important changes so far, if you're interested).

Microsoft also highlighted four new areas it's going to deal with during the rest of the year, and one of those is Windows 11's RAM usage.

Microsoft wrote: "Memory optimization for 8GB and above. Reducing Windows memory footprint to deliver a fast and responsive Windows experience across the PCs customers use every day."

As well as a promise, this is an acknowledgement of the RAM crisis, which is preventing people from upgrading their memory, but more to the point, it's forcing laptop makers to go backwards in some respects, dropping RAM loadouts to 8GB for some models. (Whereas previously we'd been moving noticeably towards 16GB as a baseline).

As to what else Microsoft has on its fixing checklist, that includes making Windows 11's setup a faster and more efficient experience, as well as a better setup for families (including parental controls being more readily accessible). Finally, Microsoft says voice functionality is also being worked on, to make "voice more natural and fluid to interact across the apps you use every day".

Analysis: it's about time really – and there's another issue that needs to be addressed, too

(Image credit: Windows/Unsplash)

Microsoft talked about RAM optimization and "lowering the baseline memory footprint for Windows" back when the campaign to fix its OS first began. So really, this is just a reassurance that this work is underway and should be arriving before the end of the year, and that it's specifically focused on ensuring 8GB is an acceptable amount of memory for running Windows 11.

As noted, there's more pressure around this now thanks to the RAM crisis getting worse, and laptop makers looking to produce more 8GB models in a cost-saving effort. On top of that, there's the embarrassment of the MacBook Neo showing that Apple can do 8GB, and do it well, with macOS.

Really, though, Windows 11 should've been better optimized for performance right from the start – or at least after the first year or two of its existence. Why has it taken a RAM apocalypse, a huge anti-Windows backlash (prompted by AI), and the launch of a rival budget-friendly MacBook to prompt Microsoft to get off its system performance laurels and actually do something serious about how well its desktop operating system runs with a leaner amount of memory? Remember, Windows 11 is supposed to work with 4GB (as a minimum), so 8GB should not be as much of an issue as it has always been.

Obviously it's good that this is now happening, and more broadly, I'm still pleased Microsoft is addressing a whole bunch of problems with its drive to fix Windows 11. There's one piece of the puzzle omitted here, though, based on some of the feedback which has come through on social media after Microsoft's blog post update.

While Redditor 'Melchett_General' acknowledges that "some of this is very nice" when it comes to Microsoft's summarized fixes, they point out some missing bits, one of which is notably echoed elsewhere – namely the recent LG monitor controversy.

The Redditor complains: "How on earth Windows got to the state where any manufacturer that makes a device can have software 'companion' applications installed silently on a user's device without the wider internet throwing a collective fit and having Microsoft permanently change course I don't know."

In case you missed it, this refers to LG serving Windows 11 users a McAfee pop-up via its monitor software, and it's a story with some degree of nuance. Microsoft did immediately take action to get LG to banish that McAfee advert, which popped up via the LG monitor app in Windows 11, but as I discussed previously, a wider problem remains here.

This is the issue that the LG application is piped through to the operating system via Windows Update without the user's knowledge, happening automatically when the LG monitor is connected to the system.

In other words, the McAfee pop-up might have been quickly dealt with, but it's still possible for hardware makers to get Windows Update to pipe through a tool for their device.

(Image credit: Future / James Holland)

As the Redditor elaborates: "What this seems to mean is that a 3rd party can install what they like whenever they like on your internet connected Windows computer and there's no effective way to stop it. (Aside from GP [Group Policy] edits or registry changes).

"This needs to be addressed, and there needs to be a mechanism to allow people to easily prevent this from happening."

I couldn't agree more. There should be no scenario where software is automatically installed in this way, even if the device maker might argue that the app is very much for the user's benefit. At the very least, there should be a dialog box announcing what's happening, and giving you the ability to choose not to install whatever application the hardware maker wants to put on your PC.

Notably, I haven't heard anything from Microsoft addressing this broader issue. Never mind the symptom (the McAfee pop-up), tackle the root cause of the issue (the ability for a piece of connected hardware to whisk its own software off the internet with no say from the user).

Microsoft is clearly serious about fixing everything that's wrong with Windows 11, I don't dispute that – but I find it odd that nothing has been said that directly addresses this pain point. Hopefully that won't be the case for much longer – we need some more transparency with Windows Update in this regard.

• Earlier this year, Microsoft confirmed that the price of Xbox consoles would rise in August, but did not reveal how much they would cost in the UK
• Now the prices are public, and the reaction has been overwhelmingly negative
• It follows a wider trend of increased prices driven by the sharp rise in the cost of RAM

Earlier this year, Microsoft confirmed that it would be raising the prices of its Xbox consoles worldwide in August, but it left the prices outside of the US a mystery. Now the price hike has actually arrived, we know how much everything now costs in the UK — and it's pretty bleak.

The price of the once entry-level 512GB Xbox Series S has shot up by £130 to an eye-watering £429.99. That's more than the newer Nintendo Switch 2 system, which retails for £395.99.

A 1TB Xbox Series X is now £669.99 (up from £499.99), meaning that there's only a £120 difference between it and the competing PlayStation 5 Pro, which costs £789.99 despite offering significantly more storage and more powerful specs.

You can see all the price changes below:

UK Xbox console pricing

Old price

New price

Xbox Series S 512GB

£299.99

£429.99

Xbox Series S 1TB

£349.99

£519.99

Xbox Series X 1TB All-Digital

£449.99

£619.99

Xbox Series X 1TB Disc

£499.99

£669.99

Here are the new prices in the US, which were announced back in June, for comparison:

US Xbox console pricing

Old price

New price

Xbox Series S 512GB

$399

$499

Xbox Series S 1TB

$449

$599

Xbox Series X 1TB All-Digital

$599

$749

Xbox Series X 1TB Disc

$649

$899

The raised prices come as the consumer electronics market is being squeezed by a global RAM shortage, which has caused the cost of key components to skyrocket. Market analysts have also said that the US-Israel war with Iran could further compound issues.

The price of other devices like laptops, phones, and tablets has also risen, but video game consoles are often sold with slim, or sometimes no, margin, which makes manufacturers particularly susceptible to rising costs.

Back in April, Sony raised the prices of its PlayStation 5 consoles by $100 / £100, citing "continued pressures in the global economic landscape". Only Nintendo has not put up the cost of its hardware since it released, though a former employee has warned that it is "inevitable."

Although part of a wider trend, the reaction to the new Xbox prices has still been overwhelmingly negative. A user on gaming forum Resetera described it as the "final nail in the coffin" for Xbox, which has struggled with sharply declining sales year-over-year.

"Makes no difference, no one is buying them either way," wrote another commenter, with others remarking on how difficult it can be to even find any Xbox products on sale in stores in Europe.

Even users that have already bought systems at lower prices seem worried: "My main concern is if a console breaks I can't afford another," one added.

An FDA advisory panel recently voted to green-light some popular peptides. The decision brought enormous attention to the market for these products — and concerns about the unregulated supply.

(Image credit: Catherine Falls Commercial)

Perhaps the issue is not that AI is creating new sprawl, but that the drive to adopt is exposing existing, disconnected processes as a deeper, more urgent problem.

The explosive growth of AI adoption across enterprises is impossible to ignore, but so is the complexity that's building alongside it. A recent study from Harvard Business Review suggests that instead of making our workloads easier, AI may actually be contributing to mental fatigue, caused by excessive interaction with AI tools beyond a person’s cognitive capacity.

It raises an important question, how can organizations manage the next wave of automation, without creating even more complexity?

The real battle aren’t your tools, it’s everything around them

There is a growing narrative that AI is SaaS's logical replacement but there’s more to it than a simple swap out. For the technology to be truly transformational, it’s about the implementation.

Through the ease of adoption, AI tools will cause more pain if simply layered on top of current systems rather than rethinking them. This isn’t a specific AI problem, it's a deployment one and organizations need to confront it by asking, “where is the real friction in the work process, and how can it be eliminated with AI?” Most people today don't realize how important that differentiation is.

The productivity gap that AI alone won’t fix

To be able to address this challenge, businesses need to properly understand the issue at hand. This starts with looking closely at where time and money is actually going, but reading through this data can be an uncomfortable task.

UK businesses are losing billions of pounds every year to “shadow work” - manual tasks outside of people’s core jobs that are exacerbated by disconnected systems. UK employees spend hours a week on this kind of non-core work, and this is only set to increase, leading to a very real problem. Every employee, team and seniority level is affected by this structural drain on output.

Whilst most businesses don’t or only partially automate tasks like planning trips and submitting expenses, these recurrent weekly friction points build up to have a quantifiable effect on the employees time for their real work which drives impact. The drive to adopt the technology is there and for leaders, automating this work should be a high priority but only when clear plans are in place for how to do so.

The fundamental problem is that shadow work emerges because systems are unable to communicate with one another, and just layering AI tools on top won't fix the problem. Often, the biggest roadblock to solving the issue in the first place is poorly integrated legacy systems.

This misalignment is draining employees productivity and satisfactions and harming business growth.

2026 - the year of stack rationalization

A true turning point may start to happen when businesses start cutting the amount of tools they have rather than adding more. For years, the cost of complexity has been gradually escalating. It's getting harder to ignore. The argument for simplicity is not just significant but essential when workers are managing shadow work, using multiple tools which are often not properly integrated.

Previously, increasing the efficiency of individual systems was the goal of enterprise automation. Payroll ran itself and IT provisioning took place in a matter of seconds. However, the coordination, approvals and administrative work that spans operations, finance and human resources were mostly ignored.

Adding more tools won't unlock the next wave of productivity. It will only come by integrating intelligence into the operational core, where people and systems come together as one.

When simplifying your tech stack is done well, it doesn't mean doing less. It involves being deliberate, putting simplicity ahead of growth and placing people before software sprawl. The CIO’s with the most self-control to base their success on what they remove will be the ones leading the next generation.

The productivity gap won't close itself

The issue for many organizations isn't the lack of technology, it’s that systems and tools are misaligned. The automation is there, but in ways that have enhanced systems without freeing up those who use them. AI won't automatically address the shadow work issue, and it won't go away on its own.

A change in perspective is now required, to move away from just pursuing the next goal and towards creating workflows that really benefit the individuals performing the task.

When complexity is outpacing productivity, the real measure of progress won’t be calculated on the volume of AI deployed within the business, but instead on how much friction an organization has managed to remove.

We've featured the best AI chatbot for business.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

Security leaders have made significant progress in improving threat visibility across businesses.

Most organizations can now identify vulnerabilities across their applications, dependencies, and development pipelines with far more consistency than previously.

Yet, a fundamental imbalance remains — vulnerabilities are being discovered faster than they can be remediated.

This imbalance is growing. As it stands, the majority (82%) of organizations currently carry security debt, categorized as accumulated vulnerabilities that have remained unresolved for more than a year.

At the same time, the share of vulnerabilities that are both severe and likely to be exploited continues to increase.

As a result, vulnerabilities are persisting in production environments long enough to be discovered and weaponized.

Despite this growing risk, many CISOs still need to convince the C-suite that reducing security debt is a business-wide issue that justifies sustained investment, rather than a challenge limited only to security teams.

Treating security debt like financial debt

Business leaders must reframe their thinking to view security debt with the same level of scrutiny as they would financial debt. Like financial debt, security debt accumulates over time, compounding when left unmanaged to create spiraling costs for the business. Those costs appear in delayed releases, emergency remediation efforts, audit findings, incident response and, ultimately, greater organizational risk.

As with financial debt, security debt requires active management rather than periodic damage control. Organizations need to clearly understand how much security debt they are carrying, distinguish between the vulnerabilities that matter most, and make deliberate decisions about where to invest their remediation efforts. Without that discipline, the backlog continues to grow while the organization's overall risk increases.

Boards already monitor financial performance, operational resilience, and service reliability because each affects the organization's ability to operate. Security debt belongs in the same category. It is a measurable indicator of organizational exposure and should be managed with the same level of oversight and accountability as any other business risk.

Capacity is the key problem, not visibility

Most organizations already know where many of their security vulnerabilities exist, however are constrained when it comes to remediation capacity. When vulnerabilities are identified faster than engineering teams can resolve them, security debt continues to grow regardless of how sophisticated the detection tools they use are.

To secure buy-in from the wider C-suite, CISOs must demonstrate this capacity gap in business terms. This includes highlighting the volume of vulnerabilities being discovered versus fixed, how long high-risk issues remain unresolved and where critical systems remain exposed.

Framing remediation as an operational constraint makes it easier for executives to understand the wider business benefits of addressing it – from improving engineering capacity to reducing costs and maintaining service availability.

Success in reducing security debt should be measured by reducing exposure, not just counting the vulnerabilities that have been found or closed. Metrics such as the number of exploitable vulnerabilities in critical systems, their average age, and overall security debt provide a clearer picture of organizational risk.

Formal risk acceptance for unresolved high-risk issues, combined with dedicated engineering time, automation, and AI-assisted remediation, can significantly improve remediation throughput without slowing development.

Prioritize the vulnerabilities that create the greatest business risk

Crucially, not every vulnerability presents the same level of risk. While severity scores such as the Common Vulnerability Scoring System (CVSS) can be useful, they do not account for exploitability, enterprise context or whether an affected application is business critical.

A more effective approach combines severity with exploitability alongside organizational context to identify the small percentage of vulnerabilities that are most likely to impact the business. Every organization has ‘crown-jewel’ applications – whether customer-facing platforms, revenue-generating services or applications handling sensitive data – and these should be prioritized for remediation.

To put this into perspective, 11% of vulnerabilities are deemed both highly severe and exploitable. Focusing resources on this subset enables organizations to reduce risk far more effectively than treating every vulnerability equally, while giving security leaders a clearer way to explain remediation priorities in business terms rather than technical jargon.

Reframing the conversation around security debt

The ripple effect of security debt extends beyond the security function. It influences resilience, regulatory compliance, and an organization's ability to operate software with confidence.

CISOs have an opportunity to reshape the conversation by framing security debt as an enterprise risk rather than a technical backlog. When leadership can understand the relationship between remediation capacity and business risk, decisions around investment and prioritization become far clearer.

Security debt will never be eradicated completely. What matters is how effectively it is measured, governed, and reduced over time. The businesses that invest in remediation capacity and focus on the vulnerabilities that matter most will be better positioned to control risk at scale.

We list the best antivirus software.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

• Microsoft Teams mobile users warned to update now
• Android and iOS users could risk losing access to their Calendar if not
• Deadline moved up to September 2026 as Microsoft likely reacts to end of life for EWS

Microsoft has warned Teams users that it has changed the date for them to update their smartphone apps, or risk losing access to the Calendar.

The company had originally said users of its Android and iOS apps would have until the end of October 2026 to update their services, but this deadline has now been brought forward to the end of September 2026.

Only the iOS and Android apps are affected, with web and desktop versions unaffected - and we would expect users with automatic updates enabled to get an updated version without needing to take any action.

Update Teams now

"To ensure continued access to Calendar functionality in Microsoft Teams mobile, users must update the Teams mobile app to the latest version," Microsoft wrote in its notice.

"This change helps maintain a reliable calendar experience on iOS and Android devices and ensures compatibility with ongoing Teams service updates."

Microsoft says those who aren't able to update (or forget, or ignore the warnings), "will not be able to access Calendar functionality in Teams mobile" - and nothing will be changing to Teams Calendar on desktop or web.

Users will need to update to version 8.9.0 (build 8.9.77.2026092302) or higher on iOS, and version 1416/1.0.0.2026122504 or higher on Android.

For those using at Teams on a work device - admins will need to make sure one of these updated versions are deployed before the deadline date.

The move appears to be linked to the upcoming end of support for Exchange Web Services (EWS) which allows apps to access mailboxes and other data in Exchange Online.

This is due to come into force on October 1, 2026, with shutdown having first been announced in 2023, following a number of high-profile security issues.

Microsoft notes that Teams admins can extend EWS until April 2027 if needed - however after that date,. it will no longer work.

Via The Register

As we enter the latter half of 2026, the AI conversation has shifted from implementation to ROI. Ninety percent of organizations now report getting some value from AI tools. Only 45% are getting a great deal of it. That gap, which is the distance between “something” and “substantial,” is where the entire AI economy is currently parked.

In Harvard Business Review earlier this year, my team and I published findings from a survey (link 1) of 1,006 C-level executives across 32 industries and 11 countries. Several factors separate the organizations getting real returns from those merely getting by. Most prominently, 55% of executives cite unready data as an inhibitor, and 47% cite the absence of a repeatable value framework.

But one variable did more to predict high value than any other: how organizations measure and report what their AI is actually worth.

We mapped that into a six-stage Economic Maturity Model. Stage 0 is unmeasured pilots. Stage 5 is formal reporting of AI value to boards, investors, and public markets. The share of organizations achieving high value climbs from 4% at Stage 0 to 85% at Stage 5, a 20x difference, and the single largest effect size in the entire study.

This is a financial measurement maturity model that measures not how well an organization builds AI, but how it measures what AI delivers.

Two cliffs, not a smooth climb

The journey between Stage 0 and Stage 5 is not linear. Plot high-value achievement against stage and two cliffs emerge, with plateaus in between.

The first cliff sits between Stage 2 and Stage 3, a 24-point jump. This is where organizations stop relying on pre-launch business cases alone and measure outcomes after deployment.

The second cliff sits between Stage 4 and Stage 5, a 27-point jump, the largest in the model. This is where organizations stop reporting AI value privately and start reporting it formally to boards and to markets. Everything between those two moves is incremental. The cliffs are where the value lives.

The 41-point trap

Thirty percent of organizations in our sample are stuck at Stage 3, with a median of six years in AI. The reason they stop is that Stage 3 feels like winning. Forty-four percent of Stage 3 organizations report a great deal of value from AI, one-third say they have substantial positive ROI, and 70% plan to increase AI investment next year.

These numbers sound like success, but at Stage 4, 58% see high value, and at stage 5, it is 85%. Stage 3 organizations are leaving 41 percentage points on the table but not feeling it, because everything is trending up from where they started.

Why do they stay stuck? Stage 3 organizations are measuring, but can’t combine what they measure into a value statement that lands with the board and the C-suite. Finance uses IRR and NPV.

Operations uses efficiency and cycle time. Risk and compliance uses incident reduction. Customer teams use CSAT and retention. Every number is real and defensible, but when the CEO or board asks, “What is our total AI value?” the answer is a shrug or four disconnected numbers from four different functions. Meanwhile the portfolios keep growing, and the translation gap scales with them.

The missing seat at the table

This brings me to the most actionable finding in our research. When the CFO and the finance function are involved in achieving and certifying AI value by partnering with the CIO, CTO, or chief data and AI officer, 76% of organizations achieve high value from AI. When the CIO or CTO carries that accountability alone, the rate drops to 53%. When functional executives carry it, 32%. Yet today, only 2% of organizations involve the CFO in this role.

The point is not to put the CFO in charge of AI, but to provide institutional credibility. Finance’s existing job is to translate different kinds of value into one common unit, and its certification of a number makes the rest of the organization believe it. The unit-CFO validation step is what makes the aggregate credible to the board, to investors and to the market.

This is also what unlocks the second cliff. As organizations move from Stage 4 to Stage 5, three things change at once, none of them technical. CIO and CTO involvement in AI value nearly doubles, from 20% to 39%. The inhibitor “our leaders don’t understand AI value” drops from 40% to 18%, because formal reporting forces executives to defend the number in public.

And investment conviction follows the proof. Just over half (52%) of Stage 5 organizations plan to substantially increase AI investment next year, which is ten times the rate at Stage 3.

Stage 4 organizations know their AI number. Stage 5 organizations are accountable for it. That shift, from private knowledge to public accountability, is what unlocks the final 27 points.

Three questions for your leadership team this week

If you take nothing else from this research, take these three:

One. Can we quantify our total AI value today, as a single number?

Two. Who owns that number — not the AI team, but an executive accountable directly to the CEO, with finance at the table?

Three. What would it take to put AI value on our board agenda as a recurring item?

If you cannot answer the first question, you are at Stage 3 or below, the same company as 62% of organizations globally. The good news is the path forward is not a technology problem, but a management decision. And it can start this quarter.

We've featured the best AI website builder.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

Dreo TurboCool Misting Fan 765S review

The Dreo TurboCool Misting Fan 765S has an advantage over traditional fans. While traditional fans provide what feels like a cooling breeze in the sweltering heat, they don’t in fact actually cool the air around you. That’s where the Dreo's clever misting technology comes in, reducing the temperature inside your home.

The brand claims the 765S can drop the temperature in a room by up to 6 degrees Celsius (roughly 11F), which makes all the difference when the mercury is rising.

It’s not the only misting fan on the market — for example the Shark FlexBreeze sits proudly atop our best fans list as I write this — but with super-quiet operation and bold room-cooling claims, can Dreo topple it from its top spot? I’ve put the 765S through its paces to find out.

Setup is pretty simple, but do take a moment before you remove the fan from its packaging. You need to remove the weighted base plate first, then prop the fan tower up on the edge of the box so you can fix the base to the bottom. It clips onto the bottom of the tower and is secured with a screw, included in the box.

There's a variety of protective packaging around the fan that needs to be removed before operation, including cardboard and plastic in the water tank. It’s worth taking your time to check you’ve removed all the various stickers, plastic film and cardboard protection — I did miss a couple of things initially.

(Image credit: Future)

Stand it up, plug it in and then setup is ready to be completed via the free Dreo app. App setup is as easy as the physical. Create an account, pair the app with the fan, enter your Wi-Fi password, and you’re good to go.

And the app is pretty good: it's clearly laid out, although it only offers basic controls (on/off and timer) on the main screen. For any deeper controls, you have to tap the picture of the fan. It’s not difficult, but I’d like to see a few more options without having to tap into another menu.

Fortunately, you're not solely reliant on the app, as there are touch controls on the top of the fan, plus you get a remote (and the two AAA batteries required to power it) included in the box.

With the remote, you can control the power, fan speed, fan mode, oscillation, misting and timer. It feels a little cheap in the hard with a basic plastic shell — Dyson’s remotes are far more premium, plus they magnetically cling to the top of the firm’s fans — but it’s functional and feels sturdy.

It’s annoying that there’s just one fan speed button on the remote though, forcing you to cycle through all speeds levels to get your desired one. I’d rather have up and down buttons for faster control.

(Image credit: Future)

When filling the water tank, it slides out from the front of the fan, and comes with a fold up carry handle to make transportation easier. You might think you don’t need it as you carry it empty to the sink, but once filled to the max level line you’ll find it surprisingly heavy. If you have a shallow sink, you might find filling a little cumbersome due to the size and shape of the tank.

The 765S's design is sleek and modern, with a pillar shape that tapers towards the top. The circular display on the front is bright and easy to read, even from across a room.

And at the bottom of the tower the water tank itself hides a fun feature. Power the fan on and you’ll notice a funky ambient light. It’s not just decorative, as the color can indicate the current humidity in the room. Blue signals humidity of 70%, while orange means the air is dry, at 30% or less. Green signals a comfortable humidity between 30% and 70%.

However, you can just have it illuminated in a color of your choice instead, with no link to the fan's performance or the room’s humidity.

There are three other lighting modes to choose from: constant, breath, and cycle. Constant allows you to pick a color of your choice for the light to be, while breath will see your color selection slowly pulsate. Cycle, as the name suggests, will cycle through a range of different colors, fading from one to the next. I found this to be a handy party mode when I had guests over and we were listening to music.

(Image credit: Future)

But what’s it like as a fan, I hear you scream? The good news is: pretty darn good.

At fan speed two, you can comfortably feel a breeze from four meters away, and it’s very quiet at this level, making it perfect for chatting with others, working, or watching TV. Even at its highest speed (12), I got a peak reading of 65dB directly in front of the machine, which is equivalent to normal conversation. It’s quieter than that when you’re sitting away, allowing you to speak to others without the fan being a distraction.

When it’s warm and you are looking for a breeze to take the edge off, this is all you really need. Higher fan speeds are useful when the fan is placed in larger, open-plan spaces, while the misting function is only really required when the mercury starts to soar.

But when the temperature does rise, the misting function is excellent. There are four misting vents — two on each side of the main airflow vent that runs down the middle — and they’re angled slightly up so moisture is directed into the air, rather than at the floor / furniture.

And it surprised me in two ways. Firstly, there’s hardly any increase in noise when the misting function is on, which means you could leave it on overnight in your bedroom and not be disturbed. Second, I was impressed with how fine the mist produced was. The ultrasonic atomizers used in the 765S produce a 17um ultra-fine mist — this means it evaporates quicker, preventing damp building up on objects nearby.

I could sit comfortably just a meter away from the fan, enjoying the cooling breeze without a sensation that it’s lightly raining. What’s even better is the misting function actually delivers on the claim that it reduces the temperature in your room.

During my tests, I saw a degree knocked off the room temperature after just 10 minutes of misting. It instantly makes a difference, and I was able to set a target humidity so the fan wouldn’t just keep pumping out water unnecessarily.

(Image credit: Future)

Keeping the fan running continuously for several hours, room temperature dipped by another degree, but I never saw more than a 2C drop during my weeks of testing. That’s some way off Dreo’s “up to 6 degrees” temperature reduction claim, and I’m not sure what sort of ideal scenario is needed to achieve this in the real world.

Still, a reduction in temperature is always a win, and the misting adds an additional layer of cooling, keeping me comfortable in what otherwise would have been a stuffy home office.

Dreo says the 6L water tank will last for 24 hours (or more) of run time when the misting mode is enabled, and I’m inclined to agree. I ran the 765S with misting turned on for a variety of different durations, and even during longer four to five hour sessions the water level reduced only marginally.

One slight frustration though was the limited oscillation angle. The 765S can only rotate 90 degrees. For many that’ll likely be enough, but having recently tested the Dyson Find+Follow Purifier Cool that has 350 degrees of movement, the Dreo feels a touch restrictive.

If you live somewhere hot and dry, the TurboCool Misting Fan 765S will provide all-day misting without breaking a sweat. Dreo claims the TurboCool Misting Fan 765S can reduce temperatures by up to 6C and although I didn’t get this level of cooling, it did make a significant difference in the rooms I tested it in, making them far more comfortable to work, relax, and sleep in.

It’s not going to deliver the same cooling as air conditioning, but the 765S offers an excellent middle ground between AC and traditional fans. It can genuinely reduce the heat in the room, keep humidity at a comfortable level, and does it all so quietly you might even forget it’s there.

(Image credit: Future)Dreo TurboCool Misting Fan 765S review: Price and release date

• The TurboCool Misting Fan 765S costs £199.99 / $199.99
• It’s available now via Amazon in the US and UK
• A cheaper, table-top version is also available

The Dreo TurboCool Misting Fan 765S is available for a list price of £199.99 / $199.99, having gone on sale in April 2026. That’s the same price as its core rival, the Shark FlexBreeze, giving it a fighting chance in the market.

If you’re after a more compact offering, Dreo also has the table-top TurboCool Misting Fan 516S for £99.99 / $99.99. Meanwhile the TurboCool Outdoor Misting Fan 711AS — also £199.99 / $199.99 at Amazon — is, as the name suggests, designed for use both indoors and out. So if you’re looking to keep things cooler whether you’re in the garden or in bed, this is the one to go for.

(Image credit: Future)Dreo TurboCool Misting Fan 765S review: Specs

Speeds

12

Modes

4

Control

Remote, app, voice, touch

Max distance

18m

Max speed

10m/s

Max air volume

3007m3/h

Sound level (Max)

65dBA (tested)

Sound level (Quiet Mode)

20dBA

Oscillation

90 degrees

Cord length

1.8m

Product diameter

280mm

Height

1,105mm

Weight

8kg

Water tank

6L

(Image credit: Future)Should I buy the Dreo TurboCool Misting Fan 765S?

Attribute

Notes

Score

Features

There’s a good range of features here, and they work well, from the quiet sleep mode and handy scheduler, to the cooling misting mode and clever illumination.

4/5

Performance

I was impressed with the quiet operation and the misting function that genuinely reduces room temperature (although not by as much as Dreo claims).

4.5/5

Design

Decent design with attractive illumination tricks that are functional and decorative. But it’s not exactly going to wow people, it’s heavy to move around (especially with a full water tank) and the remote feels a touch cheap.

4/5

Value

Competitively priced considering the feature set. If you spot it on discount, I wouldn’t hesitate to purchase it.

4.5/5

Buy it if…

You want a fan that can actually cool down your room
The misting function on the Dreo TurboCool Misting Fan 765S actually brings the temperature of your room down by a degree or two, helping you beat the heat during the day and night.

You want to sleep and keep your fan on
The TurboCool Misting Fan 765S is exceptionally quiet, and you might even forget it’s even on. You’ll easily be able to sleep with the fan set at its lower speeds, and you can even have the misting function on too.

You want hassle-free operation
The 6L water tank can offer misting abilities for over 24 hours, which means you don’t need to be constantly checking the level. Coupled with an intuitive app and smart features, the Dreo 765S is a breeze to use.

Don’t buy it if…

You’re after AC-levels of cooling
The Dreo TurboCool Misting Fan 765S does reduce the temperature of the room it's in with its clever misting feature — something a traditional fan cannot do — but it’s not going to deliver AC levels of cooling.

You struggle to move heavy objects
The 6L water tank alone is heavy enough to lift out of a sink and carry back to the fan, so the two together are pretty weighty. There is a useful carry handle on the back, but it’s a lot of weight to shift by yourself.

You want to use it in a small room
Dreo has a smaller, more affordable table-top version of the 765S that will be great for small offices or box rooms. The TurboCool Misting Fan 516S will set you back £99.99 / $99.99 — that’s half the price of the 765S.

Dreo TurboCool Misting Fan 765S review: Also consider

Shark FlexBreeze
The main competitor to the Dreo, Shark’s FlexBreeze is powerful, near-silent, and incredibly versatile. It can be used virtually anywhere — it transforms from desktop to pedestal, works corded or cordless, and is suitable for indoor and outdoor use. Its misting function is only designed to be used outdoors though, giving the Dreo the edge when it comes to indoor cooling.

Read our full Shark FlexBreeze review

Dreo TurboCool Misting Fan 516S
If you’re sold on the cooling mist of the 765S, but think it’ll be a little too large for your home, Dreo handily has a smaller version made to sit on a table, nightstand, or desk. The TurboCool Misting Fan 516S has many of the same features as its larger sibling, just in a more compact design that won’t cool as large an area.

(Image credit: Future)How I tested the Dreo TurboCool Misting Fan 765S

• Used the TurboCool Misting Fan 765S for several weeks
• Tested in rooms of varying size and temperature
• Tried out the various modes, including misting

I used the Dreo TurboCool Misting Fan 765S for several weeks in my home, testing it in rooms of varying size and temperature. I used it in my lounge to see how effectively it could cool a modest room, had it running overnight in the bedroom to experience how loud it is when sleeping, and in my open-plan kitchen/diner for large space use.

I used the included remote, the free smartphone app, and the on-device buttons to find out how easy it was to control. I also used the misting function in the day and at night to test its cooling credentials.

• Read TechRadar’s reviews guarantee
• First reviewed: July 2026

Dreo TurboCool Misting Fan 765S review

The Dreo TurboCool Misting Fan 765S has an advantage over traditional fans. While traditional fans provide what feels like a cooling breeze in the sweltering heat, they don’t in fact actually cool the air around you. That’s where the Dreo's clever misting technology comes in, reducing the temperature inside your home.

The brand claims the 765S can drop the temperature in a room by up to 6 degrees Celsius (roughly 11F), which makes all the difference when the mercury is rising.

It’s not the only misting fan on the market — for example the Shark FlexBreeze sits proudly atop our best fans list as I write this — but with super-quiet operation and bold room-cooling claims, can Dreo topple it from its top spot? I’ve put the 765S through its paces to find out.

Setup is pretty simple, but do take a moment before you remove the fan from its packaging. You need to remove the weighted base plate first, then prop the fan tower up on the edge of the box so you can fix the base to the bottom. It clips onto the bottom of the tower and is secured with a screw, included in the box.

There's a variety of protective packaging around the fan that needs to be removed before operation, including cardboard and plastic in the water tank. It’s worth taking your time to check you’ve removed all the various stickers, plastic film and cardboard protection — I did miss a couple of things initially.

(Image credit: Future)

Stand it up, plug it in and then setup is ready to be completed via the free Dreo app. App setup is as easy as the physical. Create an account, pair the app with the fan, enter your Wi-Fi password, and you’re good to go.

And the app is pretty good: it's clearly laid out, although it only offers basic controls (on/off and timer) on the main screen. For any deeper controls, you have to tap the picture of the fan. It’s not difficult, but I’d like to see a few more options without having to tap into another menu.

Fortunately, you're not solely reliant on the app, as there are touch controls on the top of the fan, plus you get a remote (and the two AAA batteries required to power it) included in the box.

With the remote, you can control the power, fan speed, fan mode, oscillation, misting and timer. It feels a little cheap in the hard with a basic plastic shell — Dyson’s remotes are far more premium, plus they magnetically cling to the top of the firm’s fans — but it’s functional and feels sturdy.

It’s annoying that there’s just one fan speed button on the remote though, forcing you to cycle through all speeds levels to get your desired one. I’d rather have up and down buttons for faster control.

(Image credit: Future)

When filling the water tank, it slides out from the front of the fan, and comes with a fold up carry handle to make transportation easier. You might think you don’t need it as you carry it empty to the sink, but once filled to the max level line you’ll find it surprisingly heavy. If you have a shallow sink, you might find filling a little cumbersome due to the size and shape of the tank.

The 765S's design is sleek and modern, with a pillar shape that tapers towards the top. The circular display on the front is bright and easy to read, even from across a room.

And at the bottom of the tower the water tank itself hides a fun feature. Power the fan on and you’ll notice a funky ambient light. It’s not just decorative, as the color can indicate the current humidity in the room. Blue signals humidity of 70%, while orange means the air is dry, at 30% or less. Green signals a comfortable humidity between 30% and 70%.

However, you can just have it illuminated in a color of your choice instead, with no link to the fan's performance or the room’s humidity.

There are three other lighting modes to choose from: constant, breath, and cycle. Constant allows you to pick a color of your choice for the light to be, while breath will see your color selection slowly pulsate. Cycle, as the name suggests, will cycle through a range of different colors, fading from one to the next. I found this to be a handy party mode when I had guests over and we were listening to music.

(Image credit: Future)

But what’s it like as a fan, I hear you scream? The good news is: pretty darn good.

At fan speed two, you can comfortably feel a breeze from four meters away, and it’s very quiet at this level, making it perfect for chatting with others, working, or watching TV. Even at its highest speed (12), I got a peak reading of 65dB directly in front of the machine, which is equivalent to normal conversation. It’s quieter than that when you’re sitting away, allowing you to speak to others without the fan being a distraction.

When it’s warm and you are looking for a breeze to take the edge off, this is all you really need. Higher fan speeds are useful when the fan is placed in larger, open-plan spaces, while the misting function is only really required when the mercury starts to soar.

But when the temperature does rise, the misting function is excellent. There are four misting vents — two on each side of the main airflow vent that runs down the middle — and they’re angled slightly up so moisture is directed into the air, rather than at the floor / furniture.

And it surprised me in two ways. Firstly, there’s hardly any increase in noise when the misting function is on, which means you could leave it on overnight in your bedroom and not be disturbed. Second, I was impressed with how fine the mist produced was. The ultrasonic atomizers used in the 765S produce a 17um ultra-fine mist — this means it evaporates quicker, preventing damp building up on objects nearby.

I could sit comfortably just a meter away from the fan, enjoying the cooling breeze without a sensation that it’s lightly raining. What’s even better is the misting function actually delivers on the claim that it reduces the temperature in your room.

During my tests, I saw a degree knocked off the room temperature after just 10 minutes of misting. It instantly makes a difference, and I was able to set a target humidity so the fan wouldn’t just keep pumping out water unnecessarily.

(Image credit: Future)

Keeping the fan running continuously for several hours, room temperature dipped by another degree, but I never saw more than a 2C drop during my weeks of testing. That’s some way off Dreo’s “up to 6 degrees” temperature reduction claim, and I’m not sure what sort of ideal scenario is needed to achieve this in the real world.

Still, a reduction in temperature is always a win, and the misting adds an additional layer of cooling, keeping me comfortable in what otherwise would have been a stuffy home office.

Dreo says the 6L water tank will last for 24 hours (or more) of run time when the misting mode is enabled, and I’m inclined to agree. I ran the 765S with misting turned on for a variety of different durations, and even during longer four to five hour sessions the water level reduced only marginally.

One slight frustration though was the limited oscillation angle. The 765S can only rotate 90 degrees. For many that’ll likely be enough, but having recently tested the Dyson Find+Follow Purifier Cool that has 350 degrees of movement, the Dreo feels a touch restrictive.

If you live somewhere hot and dry, the TurboCool Misting Fan 765S will provide all-day misting without breaking a sweat. Dreo claims the TurboCool Misting Fan 765S can reduce temperatures by up to 6C and although I didn’t get this level of cooling, it did make a significant difference in the rooms I tested it in, making them far more comfortable to work, relax, and sleep in.

It’s not going to deliver the same cooling as air conditioning, but the 765S offers an excellent middle ground between AC and traditional fans. It can genuinely reduce the heat in the room, keep humidity at a comfortable level, and does it all so quietly you might even forget it’s there.

(Image credit: Future)Dreo TurboCool Misting Fan 765S review: Price and release date

• The TurboCool Misting Fan 765S costs £199.99 / $199.99
• It’s available now via Amazon in the US and UK
• A cheaper, table-top version is also available

The Dreo TurboCool Misting Fan 765S is available for a list price of £199.99 / $199.99, having gone on sale in April 2026. That’s the same price as its core rival, the Shark FlexBreeze, giving it a fighting chance in the market.

If you’re after a more compact offering, Dreo also has the table-top TurboCool Misting Fan 516S for £99.99 / $99.99. Meanwhile the TurboCool Outdoor Misting Fan 711AS — also £199.99 / $199.99 at Amazon — is, as the name suggests, designed for use both indoors and out. So if you’re looking to keep things cooler whether you’re in the garden or in bed, this is the one to go for.

(Image credit: Future)Dreo TurboCool Misting Fan 765S review: Specs

Speeds

12

Modes

4

Control

Remote, app, voice, touch

Max distance

18m

Max speed

10m/s

Max air volume

3007m3/h

Sound level (Max)

65dBA (tested)

Sound level (Quiet Mode)

20dBA

Oscillation

90 degrees

Cord length

1.8m

Product diameter

280mm

Height

1,105mm

Weight

8kg

Water tank

6L

(Image credit: Future)Should I buy the Dreo TurboCool Misting Fan 765S?

Attribute

Notes

Score

Features

There’s a good range of features here, and they work well, from the quiet sleep mode and handy scheduler, to the cooling misting mode and clever illumination.

4/5

Performance

I was impressed with the quiet operation and the misting function that genuinely reduces room temperature (although not by as much as Dreo claims).

4.5/5

Design

Decent design with attractive illumination tricks that are functional and decorative. But it’s not exactly going to wow people, it’s heavy to move around (especially with a full water tank) and the remote feels a touch cheap.

4/5

Value

Competitively priced considering the feature set. If you spot it on discount, I wouldn’t hesitate to purchase it.

4.5/5

Buy it if…

You want a fan that can actually cool down your room
The misting function on the Dreo TurboCool Misting Fan 765S actually brings the temperature of your room down by a degree or two, helping you beat the heat during the day and night.

You want to sleep and keep your fan on
The TurboCool Misting Fan 765S is exceptionally quiet, and you might even forget it’s even on. You’ll easily be able to sleep with the fan set at its lower speeds, and you can even have the misting function on too.

You want hassle-free operation
The 6L water tank can offer misting abilities for over 24 hours, which means you don’t need to be constantly checking the level. Coupled with an intuitive app and smart features, the Dreo 765S is a breeze to use.

Don’t buy it if…

You’re after AC-levels of cooling
The Dreo TurboCool Misting Fan 765S does reduce the temperature of the room it's in with its clever misting feature — something a traditional fan cannot do — but it’s not going to deliver AC levels of cooling.

You struggle to move heavy objects
The 6L water tank alone is heavy enough to lift out of a sink and carry back to the fan, so the two together are pretty weighty. There is a useful carry handle on the back, but it’s a lot of weight to shift by yourself.

You want to use it in a small room
Dreo has a smaller, more affordable table-top version of the 765S that will be great for small offices or box rooms. The TurboCool Misting Fan 516S will set you back £99.99 / $99.99 — that’s half the price of the 765S.

Dreo TurboCool Misting Fan 765S review: Also consider

Shark FlexBreeze
The main competitor to the Dreo, Shark’s FlexBreeze is powerful, near-silent, and incredibly versatile. It can be used virtually anywhere — it transforms from desktop to pedestal, works corded or cordless, and is suitable for indoor and outdoor use. Its misting function is only designed to be used outdoors though, giving the Dreo the edge when it comes to indoor cooling.

Read our full Shark FlexBreeze review

Dreo TurboCool Misting Fan 516S
If you’re sold on the cooling mist of the 765S, but think it’ll be a little too large for your home, Dreo handily has a smaller version made to sit on a table, nightstand, or desk. The TurboCool Misting Fan 516S has many of the same features as its larger sibling, just in a more compact design that won’t cool as large an area.

(Image credit: Future)How I tested the Dreo TurboCool Misting Fan 765S

• Used the TurboCool Misting Fan 765S for several weeks
• Tested in rooms of varying size and temperature
• Tried out the various modes, including misting

I used the Dreo TurboCool Misting Fan 765S for several weeks in my home, testing it in rooms of varying size and temperature. I used it in my lounge to see how effectively it could cool a modest room, had it running overnight in the bedroom to experience how loud it is when sleeping, and in my open-plan kitchen/diner for large space use.

I used the included remote, the free smartphone app, and the on-device buttons to find out how easy it was to control. I also used the misting function in the day and at night to test its cooling credentials.

• Read TechRadar’s reviews guarantee
• First reviewed: July 2026

Xiaomi Robot Vacuum X20 Max review: one-minute review

The Xiaomi Robot Vacuum X20 Max seems like a slam dunk in the robot vacuum space. Positioned right in between the budget and upper mid range sections of the market, it delivers strong vacuum and mopping performance that make it capable of tackling a wide variety of messes in your home.

With LiDAR technology, its mapping is excellent and the robot vacuum does a great job of navigating unexpected obstacles that it might encounter during a clean. The design isn't the most exciting, though it still has a good range of features like an extendable mop arm and automatic carpet detection.

The dock is large, which might make it tricky to fit into smaller homes, but it's pretty capable as well, with hot water washing and air drying capabilities to keep the mop pads clean between uses.

It's all underpinned by a competent app that's easy to use and offers key functions such as being able to schedule cleans, restricting the robot vacuum to certain rooms, and allowing you to customize the vacuuming power and amount of water used by the mop.

The elephant in the room is that my unit developed a fault after a year of testing, occasionally spilling water during a mop pad clean. It's unfortunate and, even though it doesn't seem widespread, makes this model difficult to recommend unreservedly.

Xiaomi Robot Vacuum X20 Max review: price, range & availability

• List price $999.99 / £499.99 / AU$1,199
• Launched late 2024
• No longer available in Australia
• Limited stock in the US

The Xiaomi Robot Vacuum X20 Max has a list price of $999.99 / £499.99 / AU$1,199, and is readily available via the brand’s website in the UK. It often receives discounts too, taking the price down by £100 or so around major sales events. At this price it sits neatly between the budget and upper mid range price brackets, coming in much more expensive than under $300 / £300 / AU$500 favorites like the Roborock Q7 M5 but cheaper than the likes of the Eufy X10 Pro Omni, which goes for $799.99 / £799 / AU$1,699.95.

That latter model is a good point of comparison here, as it also features mop and similar LiDAR sensor setup. With this in mind, the Xiaomi Robot Vacuum X20 Max is very solid value - especially if you get it at that reduced price point. Accessories like replacement brushes, dust bags, and mop pads are all available via the Xiaomi website at pretty reasonable prices too.

Unfortunately, this model seems a lot harder to come by in the US. You can find it on Amazon, but it goes for an eye-watering $999.99 and stock appears limited. At this price it’s significantly worse value and you would be better off spending less on an alternative. Its list price is AU$1,199 in Australia, but it appears to no longer be available either through Xiaomi's own store, or Amazon.

• Value for money score: 3/5

Xiaomi Robot Vacuum X20 Max specs

Suction

8,000Pa

Mopping

Yes

Mop type

Dual rotating discs

Robot size (W x D)

13.78 x 3.82 inches / 350 x 97mm

Robot height

~3.82 inches / ~97mm

Dock dimensions (W x D x H)

13.39 x 16.54 x 22.17 inches / 340 x 420 x 563mm

Dust bin volume (onboard)

15.22oz / 450ml

Water tank volume (onboard)

Not stated

Dust bin volume (base)

84.5oz / 2.5l

Water tank volume (base)

135.3oz / 4l

Self-empty dust

Yes

Self-empty / refill water

Yes

Mop pad cleaning:

Yes

Wi-Fi control

Yes

Xiaomi Robot Vacuum X20 Max: design

• Raised LiDAR puck
• Extendable mop arm

The Xiaomi Robot Vacuum X20 Max isn’t pushing any boundaries when it comes to its overall design, but honestly that’s not such a bad thing.

It looks like your average robot vacuum, which is to say a big plastic disc with a chunky bumper at the front and a whole host of sensors. The most prominent of these is a big LiDAR sensor that sits in a little raised puck on the top, though there are others designed to spot obstacles or prevent the vacuum from falling off ledges.

Despite the basic look, the Xiaomi Robot Vacuum X20 Max has a bunch of useful features including two mop pads, one of which can be extended on a little arm for edge cleaning. The Xiaomi Robot Vacuum X20 Max is also capable of detecting which kind of surface it’s on, raising the mop pads when it goes over carpet to prevent any accidental stains.

Coming in at13.39 x 16.54 x 22.17 inches / 340 x 420 x 563mm, the docking station that it comes with is absolutely ginormous which could make it awkward to fit in smaller homes, but there are some good reasons for this size. Firstly, you’re getting a large 135.3oz / 4l clean water tank (removed for filling by simply popping up the lid on top of the dock) and a slightly smaller but still sizable 128.5oz / 3.8l tank for dirty water storage.

Xiaomi says that this is more than enough water to clean 2,152.8 square feet / 200 square meters on a single fill, and while my testing area was considerably smaller, I can say that it didn’t need refilling across multiple cleans.

(Image credit: Future)

The dock also has a hot water system to help wash both mop pads at 131F / 55C - a temperature that Xiaomi claims reduces odours more effectively. It dries the mop pads using hot air, a process that takes a few hours and produces a small level of sound akin to a desk fan.

There’s also a generous 84.5oz / 2.5l dust bag, which Xiaomi says only needs emptying once every 75 days in a three-person household with one pet. This seems like a very generous estimate to me, and your mileage will vary depending on your specific level of household mess. This bag size is on par with much of the competition, such as the aforementioned Eufy X10 Pro Omni, so you can expect similar performance to those models.

One feature that I particularly appreciate is the ability to remove parts of the dock for cleaning, including the little tray where the vacuum sits while charging. This isn't something that you will need to do regularly, but my previous robot vacuum tracked a decent layer of grime in its dock over the years and was comparatively pretty tricky to clean.

My only real complaint from a design standpoint is the vacuum’s incessant need to narrate what it's doing. By default it frequently declares 'start to dry the mop pad' and similar phrases in a loud, annoying voice. It's disruptive, though mercifully can be turned off in the app.

• Design score: 4/5

Xiaomi Robot Vacuum X20 Max review: performance

• Good navigation
• Deals with multiple surfaces well
• Base station pump developed issues

I've been using the Xiaomi Robot Vacuum X20 Max for a year now, and can say that it handles most daily cleaning tasks well. Its navigation capabilities are excellent and map that it generated of my cleaning area was spot on. I have the base station in the corner of an L-shape hallway and the robot never struggles to find it, even when it's been cleaning the handful of adjoining rooms.

Setting room boundaries works excellently too — if you want to restrict cleaning to a specific defined area, the vacuum won't stray beyond it. It also deals with unexpected obstacles like closed doors or moved furniture well, bumping into them once or twice and then adapting its path smartly. You don't need to worry about multiple surfaces, either. Every time I spot the vacuum going over the rug in the middle of a mop cycle, the pads are automatically raised as it detects the soft surface to prevent damage.

The vacuum picks up small debris and cleans up small spills with no sweat, and for a more intensive task I doused my kitchen floor with oats, cereal crumbs, flour, ketchup, and pretty much anything else that I could find in the cupboard.

How did the Xiaomi Robot Vacuum X20 Max fare? I sent it out on the most intensive possible cleaning mode and it made a commendable dent in the mess. The oats, crumbs, and flour were masterfully hoovered up with the exception of a few survivors in the hardest to reach corners of the room.

The ketchup was more of a challenge, and at first I was pretty worried that it wouldn't be able to overcome it: the mop pads smeared it in a thin layer over the middle of the floor at first but, sure enough, after multiple visits to the base station in order to clean itself and to refill its water tanks, it was all gone after a few runs. I did spot some red residue in the grouting between tiles, but this took just a few seconds for me to wipe up.

It's the kind of performance you would want from a mid-range robot vacuum. It's not a total replacement for doing the cleaning, but it does decrease the need for big cleans.

(Image credit: Future)

That said, my unit has developed a major issue over long-term testing. Sometimes when the vacuum cleans its own mop pads (either at the start or during a cleaning task) the motor that's meant to suck away all the dirty water doesn't kick in. This means that dirty water starts to accumulate in the base station, eventually overflowing and spilling out onto the floor.

It's frustrating (and pretty gross), and I'm struggling to resolve the issue by cleaning the water sensors, fiddling with the base station, and so on. Anecdotally, a family member bought Xiaomi Robot Vacuum X20 Max a few weeks after seeing my one, and it developed a similar fault within just a couple of months. It was, after some back and forth, repaired by Xiaomi.

Looking online, it doesn't seem to be a widespread problem, though I've still been able to find at least a few others who report the same thing. Users on Facebook and Reddit also seem to have had luck getting their vacuums replaced via Xiaomi's support service, though I suspect I can't go through this route with a review sample.

For this reason, I would recommend buying from Xiaomi directly rather than through reseller if you do chose to pick one up, as you otherwise will likely have more trouble getting support if issues do arise.

• Performance score: 2.5/5

Xiaomi Robot Vacuum X20 Max review: app

• Easy to use
• Scheduling features
• Different suction levels

The Xiaomi Robot Vacuum X20 is compatible with the Xiaomi Home app. It's easy to use and has plenty of features, including the ability to view the map generated by the robot, and its location real time. This is on top of the option to start cleans in specific rooms, schedule cleans, or clean the mop pads.

You can view the battery level, and change cleaning modes. These include basic vacuuming, mopping, vacuuming and mopping, or vacuuming before mopping. You can increase the water output too, with three levels to choose between, but bear in mind that this will increase how quickly you will need to refill everything.

I didn't encounter any bugs within the app, and it remained quick and responsive during my year of testing. The fact that you have to set up a Mi account is a bit annoying, but otherwise it offers everything you could from robot vacuum companion software.

• App score: 4.5/5

(Image credit: Future)Should you buy the Xiaomi Robot Vacuum X20 Max?

Attribute

Notes

Score

Performance

Good at vacuuming up debris and cleaning up small spills... when it works as intended.

2.5/5

Design

The Xiaomi Robot Vacuum X20 Max doesn't defy any design conventions, but it's perfectly functional and packed with features.

4/5

App

An easy to use app that offers every feature you could reasonably expect.

4.5/5

Value

Decently positioned in the UK and regularly on sale, but far too expensive if you're in the US and no longer sold in Australia..

3/5

Buy it if… 

You want a robot vacuum capable of cleaning serious mess
The Xiaomi Robot Vacuum X20 Max did a good job dealing with some serious kitchen debris and spills. It requires multiple passes for some wet stains, but it is able to get through large amounts of grime.

You have a mix of hard flooring and carpet
The Xiaomi Robot Vacuum X20 Max does a great job differentiating between different flooring types.

You need a good app
The Xiaomi Robot Vacuum X20 Max's companion app has all the features you could reasonably expect. It's a great bit of software.

Don’t buy it if… 

You need long term reliability
Although likely not widespread, it's hard to ignore the fact that my unit developed a fault after a year of use.

You can't fit a large base station
The base station here is pretty large, is an includes two water tanks in addition to the standard waste bag. Only pick this model if you know you have the space for it.

Xiaomi Robot Vacuum X20 Max review: alternatives to consider

Xiaomi Robot Vacuum X20 Max

Eufy X10 Pro Omni

Eufy Omni S1 Pro

Suction:

8,000Pa

8,000Pa

8,000Pa

Mopping:

Yes

Yes

Yes

Mop type:

Dual rotating discs

Dual rotating discs

Spinning roller

Robot size (W x D):

350×97mm

12.9 x 13.9" / 32.7 x 35.3cm

12.8 x 13.7" / 32.5 x 34.8cm

Robot height:

~97mm

4.5" / 11.4cm

3.8" / 9.7cm

Dock dimensions (W x D x H):

340×420×563mm

14.4 x 18.9 x 18.1" / 36.6 x 48 x 46cm

15.8 x 18.4 x 26.4" / 40.1 x 46.7 x 67.1cm

Dust bin volume (onboard):

450ml

Not stated

0.25L

Water tank volume (onboard):

Not stated

Not stated

0.24L

Dust bin volume (base):

2.5L

2.5L

2.5L

Water tank volume (base):

4L

Not stated

Not stated

Self-empty dust:

Yes

Yes

Yes

Self-empty / refill water:

Yes

Yes

Yes

Mop pad cleaning:

Yes

Yes

Yes

Wi-Fi control:

Yes

Yes

Yes

Eufy X10 Pro Omni
This powerful alternative from Eufy costs about the same as the Xiaomi Robot Vacuum X20 Max and offers excellent cleaning performance and a good app.

Read our full Eufy X10 Pro Omni review

Eufy Omni S1 Pro
The Omni S1 Pro, also from Eufy, is a good higher-end alternative. It's much more expensive, but comes with a raft of extra features to justify the higher price tag.

Read our full Eufy Omni S1 Pro review

How I tested the Xiaomi Robot Vacuum X20 Max

I tested the Xiaomi Robot Vacuum X20 Max in my home for over a year, using it to clean a hallway and two large rooms on a weekly basis. It cleaned a wide variety of messes in that time on a number of different surfaces, namely carpet and tiled flooring. I fully utilized the companion app in order to evaluate its performance and compared the overall experience to other robot vacuums that I own.

Read more about how we test

First reviewed August 2026

Xiaomi Robot Vacuum X20 Max review: one-minute review

The Xiaomi Robot Vacuum X20 Max seems like a slam dunk in the robot vacuum space. Positioned right in between the budget and upper mid range sections of the market, it delivers strong vacuum and mopping performance that make it capable of tackling a wide variety of messes in your home.

With LiDAR technology, its mapping is excellent and the robot vacuum does a great job of navigating unexpected obstacles that it might encounter during a clean. The design isn't the most exciting, though it still has a good range of features like an extendable mop arm and automatic carpet detection.

The dock is large, which might make it tricky to fit into smaller homes, but it's pretty capable as well, with hot water washing and air drying capabilities to keep the mop pads clean between uses.

It's all underpinned by a competent app that's easy to use and offers key functions such as being able to schedule cleans, restricting the robot vacuum to certain rooms, and allowing you to customize the vacuuming power and amount of water used by the mop.

The elephant in the room is that my unit developed a fault after a year of testing, occasionally spilling water during a mop pad clean. It's unfortunate and, even though it doesn't seem widespread, makes this model difficult to recommend unreservedly.

Xiaomi Robot Vacuum X20 Max review: price, range & availability

• List price $999.99 / £499.99 / AU$1,199
• Launched late 2024
• No longer available in Australia
• Limited stock in the US

The Xiaomi Robot Vacuum X20 Max has a list price of $999.99 / £499.99 / AU$1,199, and is readily available via the brand’s website in the UK. It often receives discounts too, taking the price down by £100 or so around major sales events. At this price it sits neatly between the budget and upper mid range price brackets, coming in much more expensive than under $300 / £300 / AU$500 favorites like the Roborock Q7 M5 but cheaper than the likes of the Eufy X10 Pro Omni, which goes for $799.99 / £799 / AU$1,699.95.

That latter model is a good point of comparison here, as it also features mop and similar LiDAR sensor setup. With this in mind, the Xiaomi Robot Vacuum X20 Max is very solid value - especially if you get it at that reduced price point. Accessories like replacement brushes, dust bags, and mop pads are all available via the Xiaomi website at pretty reasonable prices too.

Unfortunately, this model seems a lot harder to come by in the US. You can find it on Amazon, but it goes for an eye-watering $999.99 and stock appears limited. At this price it’s significantly worse value and you would be better off spending less on an alternative. Its list price is AU$1,199 in Australia, but it appears to no longer be available either through Xiaomi's own store, or Amazon.

• Value for money score: 3/5

Xiaomi Robot Vacuum X20 Max specs

Suction

8,000Pa

Mopping

Yes

Mop type

Dual rotating discs

Robot size (W x D)

13.78 x 3.82 inches / 350 x 97mm

Robot height

~3.82 inches / ~97mm

Dock dimensions (W x D x H)

13.39 x 16.54 x 22.17 inches / 340 x 420 x 563mm

Dust bin volume (onboard)

15.22oz / 450ml

Water tank volume (onboard)

Not stated

Dust bin volume (base)

84.5oz / 2.5l

Water tank volume (base)

135.3oz / 4l

Self-empty dust

Yes

Self-empty / refill water

Yes

Mop pad cleaning:

Yes

Wi-Fi control

Yes

Xiaomi Robot Vacuum X20 Max: design

• Raised LiDAR puck
• Extendable mop arm

The Xiaomi Robot Vacuum X20 Max isn’t pushing any boundaries when it comes to its overall design, but honestly that’s not such a bad thing.

It looks like your average robot vacuum, which is to say a big plastic disc with a chunky bumper at the front and a whole host of sensors. The most prominent of these is a big LiDAR sensor that sits in a little raised puck on the top, though there are others designed to spot obstacles or prevent the vacuum from falling off ledges.

Despite the basic look, the Xiaomi Robot Vacuum X20 Max has a bunch of useful features including two mop pads, one of which can be extended on a little arm for edge cleaning. The Xiaomi Robot Vacuum X20 Max is also capable of detecting which kind of surface it’s on, raising the mop pads when it goes over carpet to prevent any accidental stains.

Coming in at13.39 x 16.54 x 22.17 inches / 340 x 420 x 563mm, the docking station that it comes with is absolutely ginormous which could make it awkward to fit in smaller homes, but there are some good reasons for this size. Firstly, you’re getting a large 135.3oz / 4l clean water tank (removed for filling by simply popping up the lid on top of the dock) and a slightly smaller but still sizable 128.5oz / 3.8l tank for dirty water storage.

Xiaomi says that this is more than enough water to clean 2,152.8 square feet / 200 square meters on a single fill, and while my testing area was considerably smaller, I can say that it didn’t need refilling across multiple cleans.

(Image credit: Future)

The dock also has a hot water system to help wash both mop pads at 131F / 55C - a temperature that Xiaomi claims reduces odours more effectively. It dries the mop pads using hot air, a process that takes a few hours and produces a small level of sound akin to a desk fan.

There’s also a generous 84.5oz / 2.5l dust bag, which Xiaomi says only needs emptying once every 75 days in a three-person household with one pet. This seems like a very generous estimate to me, and your mileage will vary depending on your specific level of household mess. This bag size is on par with much of the competition, such as the aforementioned Eufy X10 Pro Omni, so you can expect similar performance to those models.

One feature that I particularly appreciate is the ability to remove parts of the dock for cleaning, including the little tray where the vacuum sits while charging. This isn't something that you will need to do regularly, but my previous robot vacuum tracked a decent layer of grime in its dock over the years and was comparatively pretty tricky to clean.

My only real complaint from a design standpoint is the vacuum’s incessant need to narrate what it's doing. By default it frequently declares 'start to dry the mop pad' and similar phrases in a loud, annoying voice. It's disruptive, though mercifully can be turned off in the app.

• Design score: 4/5

Xiaomi Robot Vacuum X20 Max review: performance

• Good navigation
• Deals with multiple surfaces well
• Base station pump developed issues

I've been using the Xiaomi Robot Vacuum X20 Max for a year now, and can say that it handles most daily cleaning tasks well. Its navigation capabilities are excellent and map that it generated of my cleaning area was spot on. I have the base station in the corner of an L-shape hallway and the robot never struggles to find it, even when it's been cleaning the handful of adjoining rooms.

Setting room boundaries works excellently too — if you want to restrict cleaning to a specific defined area, the vacuum won't stray beyond it. It also deals with unexpected obstacles like closed doors or moved furniture well, bumping into them once or twice and then adapting its path smartly. You don't need to worry about multiple surfaces, either. Every time I spot the vacuum going over the rug in the middle of a mop cycle, the pads are automatically raised as it detects the soft surface to prevent damage.

The vacuum picks up small debris and cleans up small spills with no sweat, and for a more intensive task I doused my kitchen floor with oats, cereal crumbs, flour, ketchup, and pretty much anything else that I could find in the cupboard.

How did the Xiaomi Robot Vacuum X20 Max fare? I sent it out on the most intensive possible cleaning mode and it made a commendable dent in the mess. The oats, crumbs, and flour were masterfully hoovered up with the exception of a few survivors in the hardest to reach corners of the room.

The ketchup was more of a challenge, and at first I was pretty worried that it wouldn't be able to overcome it: the mop pads smeared it in a thin layer over the middle of the floor at first but, sure enough, after multiple visits to the base station in order to clean itself and to refill its water tanks, it was all gone after a few runs. I did spot some red residue in the grouting between tiles, but this took just a few seconds for me to wipe up.

It's the kind of performance you would want from a mid-range robot vacuum. It's not a total replacement for doing the cleaning, but it does decrease the need for big cleans.

(Image credit: Future)

That said, my unit has developed a major issue over long-term testing. Sometimes when the vacuum cleans its own mop pads (either at the start or during a cleaning task) the motor that's meant to suck away all the dirty water doesn't kick in. This means that dirty water starts to accumulate in the base station, eventually overflowing and spilling out onto the floor.

It's frustrating (and pretty gross), and I'm struggling to resolve the issue by cleaning the water sensors, fiddling with the base station, and so on. Anecdotally, a family member bought Xiaomi Robot Vacuum X20 Max a few weeks after seeing my one, and it developed a similar fault within just a couple of months. It was, after some back and forth, repaired by Xiaomi.

Looking online, it doesn't seem to be a widespread problem, though I've still been able to find at least a few others who report the same thing. Users on Facebook and Reddit also seem to have had luck getting their vacuums replaced via Xiaomi's support service, though I suspect I can't go through this route with a review sample.

For this reason, I would recommend buying from Xiaomi directly rather than through reseller if you do chose to pick one up, as you otherwise will likely have more trouble getting support if issues do arise.

• Performance score: 2.5/5

Xiaomi Robot Vacuum X20 Max review: app

• Easy to use
• Scheduling features
• Different suction levels

The Xiaomi Robot Vacuum X20 is compatible with the Xiaomi Home app. It's easy to use and has plenty of features, including the ability to view the map generated by the robot, and its location real time. This is on top of the option to start cleans in specific rooms, schedule cleans, or clean the mop pads.

You can view the battery level, and change cleaning modes. These include basic vacuuming, mopping, vacuuming and mopping, or vacuuming before mopping. You can increase the water output too, with three levels to choose between, but bear in mind that this will increase how quickly you will need to refill everything.

I didn't encounter any bugs within the app, and it remained quick and responsive during my year of testing. The fact that you have to set up a Mi account is a bit annoying, but otherwise it offers everything you could from robot vacuum companion software.

• App score: 4.5/5

(Image credit: Future)Should you buy the Xiaomi Robot Vacuum X20 Max?

Attribute

Notes

Score

Performance

Good at vacuuming up debris and cleaning up small spills... when it works as intended.

2.5/5

Design

The Xiaomi Robot Vacuum X20 Max doesn't defy any design conventions, but it's perfectly functional and packed with features.

4/5

App

An easy to use app that offers every feature you could reasonably expect.

4.5/5

Value

Decently positioned in the UK and regularly on sale, but far too expensive if you're in the US and no longer sold in Australia..

3/5

Buy it if… 

You want a robot vacuum capable of cleaning serious mess
The Xiaomi Robot Vacuum X20 Max did a good job dealing with some serious kitchen debris and spills. It requires multiple passes for some wet stains, but it is able to get through large amounts of grime.

You have a mix of hard flooring and carpet
The Xiaomi Robot Vacuum X20 Max does a great job differentiating between different flooring types.

You need a good app
The Xiaomi Robot Vacuum X20 Max's companion app has all the features you could reasonably expect. It's a great bit of software.

Don’t buy it if… 

You need long term reliability
Although likely not widespread, it's hard to ignore the fact that my unit developed a fault after a year of use.

You can't fit a large base station
The base station here is pretty large, is an includes two water tanks in addition to the standard waste bag. Only pick this model if you know you have the space for it.

Xiaomi Robot Vacuum X20 Max review: alternatives to consider

Xiaomi Robot Vacuum X20 Max

Eufy X10 Pro Omni

Eufy Omni S1 Pro

Suction:

8,000Pa

8,000Pa

8,000Pa

Mopping:

Yes

Yes

Yes

Mop type:

Dual rotating discs

Dual rotating discs

Spinning roller

Robot size (W x D):

350×97mm

12.9 x 13.9" / 32.7 x 35.3cm

12.8 x 13.7" / 32.5 x 34.8cm

Robot height:

~97mm

4.5" / 11.4cm

3.8" / 9.7cm

Dock dimensions (W x D x H):

340×420×563mm

14.4 x 18.9 x 18.1" / 36.6 x 48 x 46cm

15.8 x 18.4 x 26.4" / 40.1 x 46.7 x 67.1cm

Dust bin volume (onboard):

450ml

Not stated

0.25L

Water tank volume (onboard):

Not stated

Not stated

0.24L

Dust bin volume (base):

2.5L

2.5L

2.5L

Water tank volume (base):

4L

Not stated

Not stated

Self-empty dust:

Yes

Yes

Yes

Self-empty / refill water:

Yes

Yes

Yes

Mop pad cleaning:

Yes

Yes

Yes

Wi-Fi control:

Yes

Yes

Yes

Eufy X10 Pro Omni
This powerful alternative from Eufy costs about the same as the Xiaomi Robot Vacuum X20 Max and offers excellent cleaning performance and a good app.

Read our full Eufy X10 Pro Omni review

Eufy Omni S1 Pro
The Omni S1 Pro, also from Eufy, is a good higher-end alternative. It's much more expensive, but comes with a raft of extra features to justify the higher price tag.

Read our full Eufy Omni S1 Pro review

How I tested the Xiaomi Robot Vacuum X20 Max

I tested the Xiaomi Robot Vacuum X20 Max in my home for over a year, using it to clean a hallway and two large rooms on a weekly basis. It cleaned a wide variety of messes in that time on a number of different surfaces, namely carpet and tiled flooring. I fully utilized the companion app in order to evaluate its performance and compared the overall experience to other robot vacuums that I own.

Read more about how we test

First reviewed August 2026

• The InkLeaf is a notebook-like dual-screen ereader
• It opens like a book and lets you read on one screen while making notes on the other, using a stylus
• It's shipping to early buyers this month at a hefty discount

In recent years we’ve seen all sorts of e-paper innovations, from e-paper phones like the Hisense A10 to absolutely tiny ereaders like the Xteink X4 Pro, and now another unusual device is on the way, dubbed the InkLeaf.

This is a dual-screen ereader that folds open like a notebook. So the design is a bit like that of the Samsung Galaxy Z Fold 8 Ultra and other foldable phones, except it’s using E Ink for its displays, and it’s not using foldable screens.

Instead, you get a pair of 10.3-inch E Ink Carta displays, and an aluminum frame between them. That also means the screens are much bigger than on Samsung's foldable, putting each display more in line with a tablet.

The idea here is that you can hold and fold it like a notebook, and that, thanks to having two screens, you can read from one and take notes on the other. For example, you might have a textbook open on one display and note paper on the other — though you can also directly annotate whatever text you’re viewing too.

The InkLeaf, then, is designed primarily for work, study, and research, though given that it opens and folds like a book, it could also be appealing to read novels on for anyone who otherwise prefers hardbacks.

The InkLeaf is also just 4.2mm thick when open and 8.4mm thick when closed, so it’s no thicker than a notebook, and it weighs just 480g. Plus it has a speaker, microphones, 64GB of storage, and a battery that promises weeks of life per charge.

Unclear software and a high price

(Image credit: InkLeaf)

The InkLeaf runs Android under the hood, though it’s not clear what version of Android this is, or whether it includes access to the Google Play Store. It also supports PDF, EPUB, MOBI, TXT, and HTML files.

The InkLeaf comes with a stylus for all your note-taking needs, supporting 4096 levels of pressure, and this supposedly provides the friction of real paper, so it won’t feel like you’re writing on a screen.

And, of course, this being 2026, the InkLeaf also includes AI, allowing you to search for handwritten phrases across all of your notes, and with summaries and an outline assist planned post-launch.

At full price, this ereader isn’t cheap, with an MSRP of $999 (around £740 / AU$1,425), but you can reserve one now for delivery worldwide on InkLeaf's site for just $489 (roughly £365 / AU$700) — that’s the ‘founder price’, and requires paying a refundable $9.99 (approximately £7.50 / AU$14) deposit ahead of when it ships sometime this month.

As well as getting you a big discount, this is also an early access launch — following that, it will land on Kickstarter, with most units not shipping until 2027.

So if you want the InkLeaf early and affordably, you should reserve one soon. Of course, doing so comes with all the perils of being an early adopter, and means you won’t get a chance to read reviews before making a purchase decision.

Enterprise AI today operates on two distinct levels. Horizontal AI addresses common business processes broadly: useful, scalable, and increasingly commoditized.

Vertical AI goes deeper, solving complex, industry-specific challenges by integrating domain expertise, proprietary data, and core workflows.

Understanding the distinction between the two is what separates incremental efficiency gains from genuine competitive advantage.

For years, enterprise leaders have treated AI like a silver bullet. Deploy a digital assistant here, a copilot there, and watch the magic happen.

Except the magic isn’t happening, at least not at the scale the business needs.

Horizontal solutions are easy to implement, which is precisely why they’re becoming commoditized. And when everyone has the same tool, nobody has a competitive advantage.

Here’s what I’ve observed across our customer base: the enterprises pulling ahead today aren’t abandoning horizontal AI. They’re building on top of it.

They’re the ones who stopped treating broad deployment as the destination and started architecting vertical AI systems that fundamentally reshape how their business operates.

Horizontal is the foundation. Vertical is where the competitive advantage lives.

The Complexity Paradox

There’s a paradox at play that most organizations haven’t grasped yet. Horizontal AI is simple to deploy but delivers marginal value in isolation. Vertical AI is harder to build but transforms your P&L. Most enterprises are getting stuck at the first layer without progressing to the second.

Think about it this way: a generic digital assistant might save your customer service team a few hours per week. That’s real, and it’s a necessary starting point, but it’s not transformative on its own.

Now contrast that with an AI system specifically engineered for your industry, integrated into your core workflows, and leveraging your proprietary data and domain expertise. That’s not an assistant — that’s business process engineering. The two approaches work best in combination, not in competition.

The difference isn’t academic. In our workshops across Europe I’ve watched organizations that layered vertical AI capabilities onto their existing horizontal foundations unlock productivity gains of 15 to 20 percent or more, not per function but across entire workflows. The horizontal layer created the connective tissue; the vertical layer created the step change.

Why Vertical Elevates Everything

Vertical AI elevates what horizontal AI has already put in place. It does something that broad solutions alone cannot: it understands your business. It knows your processes, your constraints, your compliance requirements, and your revenue drivers. It’s not trying to be everything to everyone — it’s purpose-built for your specific problem, running on top of the platforms and workflows your teams already use.

This is where the ecosystem model takes over. The cloud infrastructure world followed the same pattern: hyperscalers provided the foundational layer, specialized vendors built on top, and system integrators orchestrated everything together. Enterprise AI is replicating that logic right now. Rather than spending months evaluating generic point solutions, organizations can now discover, test, and deploy industry-specific agents in days, sitting on top of the systems they already have.

The most significant recent shift is that enterprise platforms themselves are now shipping coordinated teams of specialized agents, purpose-built for outcomes like cash collection, workforce scheduling, or supplier sourcing, rather than leaving organizations to bolt AI onto the side of their existing systems.

This is no longer theoretical; it’s operational. We’re seeing system integrators and independent business software vendors delivering validated, vertical AI solutions across finance, HR, supply chain, and customer experience. These aren’t generic tools replacing existing capabilities but intelligent agents purpose-built for specific industries and functional challenges, amplifying the value of the platforms already in place.

The Real Risk: Stopping at the First Layer

Here’s what worries me more than anything: the organizations that are comfortable with their horizontal AI deployments and treat them as the finish line. They’ve ticked a box, deployed AI, and can talk about it in earnings calls. But they’ve fundamentally misunderstood the inflection point we’re at.

Competitive advantage in AI used to mean having access to the best talent or the biggest budget, but that’s no longer sufficient. It means having the clarity to move beyond commodity solutions and the discipline to invest in vertical capabilities that drive real business impact. The organizations doing this now are building a structural advantage that is genuinely defensible.

By contrast, enterprises that treat horizontal deployment as the endgame are already finding themselves with capable but ultimately marginal returns, and beginning to wonder why competitors are operating at fundamentally different economics even though they started from the same baseline.

The Path Forward

The message is not to abandon horizontal AI. It’s to sequence your investments correctly. Start with the embedded capabilities already in your systems. Use them to drive efficiency, build confidence, and establish the data foundations you’ll need. Then immediately follow with vertical AI projects that address your highest value, most complex business processes.

Get serious about which processes matter most to your P&L. Build or partner to develop AI systems specifically engineered for those processes. Measure relentlessly. And then scale what works.

The enterprises pulling ahead aren’t the ones with the most AI projects. They’re the ones with the most disciplined approach to both layers, understanding that horizontal AI opens the door and vertical AI is what you build once you’re through it. Complexity isn’t a barrier to deployment anymore. It’s the path to real competitive advantage.

We've reviewed, rated, and ranked the best business laptops.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

Enterprise DevOps teams have adopted AI faster than they have adapted the operational systems that support it. Large language models generate Infrastructure as Code, deployment pipelines, and configuration files in seconds.

Development teams are delivering infrastructure changes at a pace that would have been implausible five years ago. Production environments have simultaneously become harder to reproduce, more difficult to govern, and more expensive to operate.

The problem is easy to miss at first, because AI looks like a clear win in the early stages. A team asks an AI tool to set up some infrastructure. It works. The team expands how much they rely on it. More teams follow. More infrastructure gets created for testing, demos, experiments, and development. The organization ends up with far more moving parts than it had before, but far less ability to track them.

Not the full picture

The tools generating this infrastructure are good at describing what something should look like at the moment it's created. They're not built to watch what happens to it afterward, decide whether it's safe to make another change while one is already in progress, or notice when something has quietly drifted out of its intended state over the following weeks.

That kind of oversight used to come from experienced operations people making changes one at a time, in a predictable order, with a clear sense of what else was happening around them. AI removed the slow part of that process without replacing the judgment that made it manageable.

Now there are multiple people and multiple AI tools all making changes to the same systems around the same time, each one doing its job correctly in isolation, none of them able to see the full picture of what's happening across the environment as a whole.

The result looks familiar to anyone running infrastructure today. Environments that worked perfectly last week suddenly don't, and nobody can say exactly why. Cloud costs creep up steadily because nobody remembered to shut something down once it was no longer needed. Teams spend real, billable hours figuring out why two environments that were supposed to be identical have quietly diverged.

Security settings that were correct at launch slowly fall out of sync as everything around them keeps evolving independently. None of these are dramatic failures on their own. They're slow leaks, and slow leaks are much harder to notice and much more expensive to fix than sudden breaks.

Too much fragmentation

The teams responsible for infrastructure have been adjusting to this reality in real time. A few years ago, their main job was making things easy for developers, such as removing friction, giving people self-service access to what they needed, getting out of the way. That's still part of the job, but the role has expanded considerably.

Now they're expected to enforce rules across the organization, manage costs on increasingly expensive compute, track what's actually running where, and keep things consistent across a far more complicated set of environments than existed even two or three years ago. Bolting on another automated tool to handle some piece of that doesn't solve the underlying issue.

More often than not, it just adds another piece to a pile that's already too fragmented to manage well.

An environment isn't just the infrastructure sitting inside it. It's also everything connected to that infrastructure: who has access to what, what depends on what else, how much it actually costs to keep running, and how long it's supposed to stick around before someone should be cleaning it up.

Getting ahead of problems means building that kind of awareness directly into how infrastructure gets deployed and maintained, rather than discovering issues after the fact through documentation nobody reads regularly or incidents nobody wanted to have in the first place.

Processes should be built in

This becomes even more important as companies pour money into AI-specific hardware. Setting up GPU-powered environments isn't simply a matter of securing the hardware, it requires everything around that hardware, including networking, storage, security, and software, to come together correctly and stay that way.

Having the hardware on hand doesn't make that coordination problem any smaller. Only better processes do, and that process has to be built deliberately rather than assembled from whatever tools happened to be lying around.

The questions worth asking have changed as a result, even if a lot of organizations haven't caught up to that yet. How fast a team can generate infrastructure isn't the most useful thing to measure anymore and it was never really the hard part.

A better gauge is whether environments stay consistent over time, how often things drift without anyone noticing, how long unused environments sit around quietly costing money, and whether anyone in the organization can actually describe with confidence what's running in production at any given moment.

The companies that get the most lasting value out of AI won't necessarily be the ones using the most advanced or most numerous tools.

They'll be the ones that put just as much disciplined effort into managing what AI produces as they did into producing it in the first place and treating governance not as an afterthought, but as the thing that determines whether all that speed actually turns into something worth having.

We've featured the best laptop for programming.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

Tens of millions of Americans have criminal records that can be barriers to employment. An AI-powered app is helping those eligible to expunge their records at a pace not seen before.

(Image credit: Meredith Rizzo for NPR)

Kim Aris, son of Aung San Suu Kyi, told NPR that photographs of his mother meeting a foreign official gave him "a reason to hope" for the first time since the military coup that ousted her in 2021.

(Image credit: Myanmar President Office)

“AI readiness” tends to be used as umbrella phrase to describe the different range of activities that businesses need to carry out to get value of generative AI and minimize any associated risks.

AI readiness therefore takes into account everything from having the right data foundations and governance in place to supporting AI literacy among staff to establishing ethical guardrails.

One area where businesses – including even large brands – are collectively not ready for AI is ensuring their brand is visible and faithfully represented across Large Language Models (LLMs) and in responses from associated AI tools like ChatGPT, Gemini and Claude.

The outside-in view

Too many large companies have a digital estate that contains content that is outdated or contradictory. Because AI tools base their responses on the content they find, brands risk AI tools giving out misinformation or misrepresenting their brand, products or organization.

Similarly, there is also a risk of brands not being featured in responses or less than those compared to competitors, impacting future visibility and competitiveness.

This is effectively AI looking from the “outside in” and then presenting a picture of what it has surfaced. When we measured this “outside in” element of AI readiness across the websites of over 250 top global brands using an automated remote website assessment tool and an underlying AI maturity model, we found only approximately 3% of websites could be considered “leading” when it comes to AI readiness.

Why AI readiness is important

The way people search for information on the web is changing rapidly. Consumers are turning to ChatGPT rather than using traditional Google searches; surveys have suggested that 37% of consumers in the US who already use AI, now start their searches using AI tools rather than Google.

This means that rather than looking at a web page on a company’s own website, consumers’ perceptions about a brand are increasingly coming from AI responses as the first step rather than directly from a web page controlled by the company.

That’s a problem if a consumer asks AI to provide a list of suppliers or vendors, and your business isn’t listed when it should be, or the AI provides erroneous information about your company and products from the off.

Recognizing the speed at which consumers are turning to AI tools for search, digital marketing teams who have previously invested in Search Engine Optimization (SEO) are now investing in related practices like Generative Engine Optimization (GEO) and Answers Engine Optimization (AEO) with some urgency to optimize their digital footprint so it is successfully reflected in AI tools like ChatGPT.

But the reasons that even some of the world’s top brands still have a lot of work to do in this area is because in the past they have deprioritized basic governance and maintenance across their digital estate, with the risk of AI confidently presenting answers as facts or broadcasting perceptions that are based on information from content that is out of date and has been superseded.

Of course, the risks associated with erroneous answers may vary but they can be acute, particularly in heavily regulated industries such as financial services, healthcare or pharmaceuticals with strict compliance surrounding the information that businesses must provide to customers, or in areas such as products where health and safety instructions must be correct.

The hidden digital footprint

One of the key reasons where businesses slip up is by building up a digital footprint – often without realizing it – which is not actively managed and has old and inaccurate information.

Over time, businesses accumulate a surprisingly large number of digital assets that are accessible online, sometimes hosted on third-party platforms: websites for sub-brands or locations, online newsletters, campaigns, product launches, events, microsites, partner content, customer portals, forms, digital experiments, and so forth. In addition, there might be areas of a corporate website that have effectively been forgotten about.

Over the years, this digital footprint in some organizations has not been properly managed, so a site created to support a one-off event stays online indefinitely rather than being retired. Sometimes when a company acquires another, it also becomes responsible for this digital debris that has accumulated and which the new team have little awareness of the full footprint they have inherited.

Previous research we carried out between 2017 and 2023 shows that central digital teams can be unaware of up to 41% of their entire digital footprint.

The danger of poor document management

Another critical cause of a lack of AI readiness is an overreliance on documents. Many websites contain PDF attachments which don’t necessarily get updated and then accumulate over years and years, containing information which is no longer relevant or correct.

Annual reports, brochures, manual forms, product guides and more are all left accessible on websites. Our research suggests that as many as one in five (19%) website documents are duplicates which can create inconsistent signals for AI about which is the current and authoritative version

Not enough organizations take a structured approach to managing their documents across their digital estate, which means they cannot effectively update documentation for example to reflect brand, regulatory or operational changes, with AI now forming responses from these out-of-date files.

How to reduce risk exposure

As more businesses find AI is producing misinformation about them and the number of consumers who use AI tools for searching information increases, awareness of this problem is likely to rise.

To reduce the risk exposure, at a high-level, three things generally need to happen:

• The issue needs to be owned at the executive or board level with someone responsible to tackle it.
• Digital teams need to use automated assessment tools to measure and monitor the risk associated with their entire digital estate in order to track associated remedial action.
• The digital team need to put governance and lifecycle management processes in place that enables them to have more control and oversight over their digital footprint, so it avoids becoming a digital wild west again.

AI readiness covers many areas. A key priority for brands who want to maximize AI visibility while reducing the chance of misinformation is to get their digital estate under control. Of course, there are other important activities to consider including ensuring content is optimally structured, but reducing the amount of out-of-date content is a critical starting point.

We've featured the best AI chatbot for business.

This article was produced as part of TechRadar Pro Perspectives, our channel to 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/pro/perspectives-how-to-submit

SNAP, also known as food stamps, has seen a rapid and consistent participation decline since last July. The drop comes as the program faces historic cuts and changes.

(Image credit: Joe Raedle)

The Justice Department is using federal conspiracy charges to sweep up ICE protesters. Some legal experts worry the agency is casting too wide a net to prosecute political activists who oppose the Trump administration's deportation agenda.

(Image credit: Jaida Grey Eagle for NPR)