I wish generative AI (genAI) tools were truly useful. They’re not. I keep tinkering with the programs — ChatGPT, Meta AI, Gemini, etc., etc. Mind you, they look useful if you don’t know any better. Their answers sound plausible. But if you look closer, even if you forgive them for their hallucinations — that is, lies — you’ll see all too often that the answers they give are wrong.
If you’re operating at, say, a high-school-grade report level, genAI answers are fine. (Sorry, teachers.) But if you’re digging deep into a subject, which is where I live, it’s another story.
I know more than the average large language model (LLM) about subjects such as Linux and open-source software. What genAI can tell you about those subjects might sound right, but the deeper you dive into the details, the poorer the information.
Indeed, all too often, the end result is annoying and obnoxiously wrong. Worse still, it’s erratically wrong. If I could count on its answers being mediocre, but reasonably accurate, I could work around it. I can’t.
In my book, even the best genAI chatbot, Perplexity, seems to be going downhill. One of the things I liked about Perplexity was that it would give me sources for its observations. I found it useful when I could think of it as a turbo-powered search engine. In the last few days, though, it’s been getting flaky, and I can no longer count on it pointing out where it found the information it repackaged for me.
Why is this happening? I’m not an AI developer, but I pay close attention to the field and see at least two major reasons they’re beginning to fail.
The first is the quality of the content used to create the major LLMs has never been that good. Many include material from such “quality” websites as Twitter, Reddit, and 4Chan.
Unless you glue rocks into your pizza, those are silly, harmless examples, but if you need the right answer, it’s another matter entirely. Take, for example, the lawyer whose legal paperwork included information from AI-made-up cases. The judges were not amused.
If you want to sex chat with genAI tools, which appears to be one of the most popular uses for ChatGPT, accuracy probably doesn’t matter that much to you. Getting the right answers, though, is what matters to me and should matter to anyone who wants to use AI for business.
That leads to the second problem. Today, more than ever, genAI-produced content is replacing expert human content. The result isn’t just that the bad data pushes out the good; it’s more insidious than that.
A recent Nature paper found that “indiscriminately learning from data produced by other models causes ”model collapse” — a degenerative process where, over time, models forget the true underlying data distribution, even in the absence of a shift in the distribution over time.” I just like to call this Garbage In, Garbage Out (GIGO).
Since I see no chance that companies will stop using genAI to produce documents for a fraction of the cost of real subject matter experts and experienced writers, quality will only continue downhill.
We might, believe it or not, already be at peak AI as far as quality goes. Isn’t that a scary thought?
But as a way to replace knowledge workers, which all too many companies seem intent upon, forget about it. Except for CEOs — they could be replaced by AI chatbots. I doubt many people would notice much of a difference — except it would save companies a ton of money.
As the EU does its deluded best to turn iOS into Android, the Apple ecosystem is focused on what really matters to business users: security, privacy, and platform integrity. That’s why enterprise tech leaders need to know that Apple MDM and security vendor Jamf has got even closer to leading ID tech purveyor Okta.
Jamf and Okta have worked together for years and in 2023 became the first to support Apple’s Single Sign-On (SSO) framework on Macs, following a similar introduction on iPhones and iPads shortly before.
Identity, management, and security
What’s new is that Jamf has joined Okta’s Elevate partners program, meaning the two firms will be able to work even more closely together on offering combined services to enterprise customers, while also building new service offerings.
Henry Patel, Chief Strategy Officer at Jamf, explained: “With Okta as the identity provider and Jamf as the management and security solution, joint customers can offer their end users uninterrupted and productive workflows, anywhere and anytime.”
The point of this combination won’t be lost on IT.
Jamf can deliver the MDM features Apple fleets need, while also protecting against security threats, spanning all the way from the endpoint to the enterprise core. The addition of core support for Okta’s rapidly evolving collection of tools to protect and manage identity gives customers business-class management and protection of their device fleets.
What the joint solution provides
As the companies put it, their joint solution lets customers handle all the following tasks:
Enhance security with multifactor authentication (MFA) and passwordless single sign-on (PSSO).
Enable seamless device enrollment with Enrollment SSO.
Improve productivity with uninterrupted workflows, requiring fewer sign-ins with PSSO. No one likes to enter logins more often than they need to; it damages the user experience and rattles concentration.
Ensure continuous conditional access, adjusting user access permissions based on changes in security status.
Provide easy access to company resources and apps.
Conditional access is particularly important in modern working environments.
Managing complex remote work environments
While there remain some well publicized holdouts against the distributed future of hybrid working, the Bring Your Own Device (BYOD) trend showed the inevitability of workplace changes over time.
The problem with that scenario is that enterprises must figure out how to protect their data outside of traditional perimeter security models, and that’s what conditional access tries to do. It’s a system that relies on signals such as location, time, device, or user to generate an insight into the extent to which a device can be trusted when it is used to try to gain access to company data.
There’s a mountain of work that has taken place around conditional access across the last few years, and this task accelerated during the pandemic. Apple’s Declarative Device Management tech, Jamf support for Microsoft Intune, and inclusion of Jamf at the Azure Marketplace are all testament to the degree to which tech is building rock-solid identity and access solutions, while SSO and passkeys show the user focus inherent to all these attempts.
Enterprise-class simplicity
Ideally, I suppose, device and endpoint security would be something that happened without any user involvement. (People tend to be the weakest link in security, after all.) And it is good to see that kind of future being realized across Macs, iPhones, and iPads.
Now, having said that even SSO shows there will always be some need for user interaction but reading between the lines of the Jamf/Okta announcement, it’s hard not to detect the intention: to deliver enterprise-grade security that’s simple and easy to use just like any of Apple’s own consumer technologies, protection that enables workers to focus on what they do, while giving IT the degree of control, security and identity provision they require. All of this delivered without an ounce of kernel access — like an enterprise-class platform done right, some might say.
Ah, the ever-present annual conundrum for any Pixel-appreciating Android aficionado: To upgrade or not to upgrade?
Google’s latest Pixel 9 devices are here earlier than ever this year, and while they may not feel like night-and-day departures from the previous Pixel models, they pack plenty of interesting improvements.
It’s certainly enough to be tempting — and, potentially, to be a truly significant step up from your current Google-made Pixel gadget. But is it actually worth your while to get a Pixel 9, or are you better off hanging onto your current Pixel phone for a while longer?
Having lived with the Google Pixel 9, Pixel 9 Pro, and Pixel 9 Pro XL for a full week now, thanks to review units provided on loan from Google — and coming from the perspective of someone who personally owns a Pixel 8 Pro, has owned plenty of Pixels before that, and has spent a significant amount of time with every single Pixel model to date — lemme tell ya: There isn’t a simple, one-size-fits-all answer.
But there is some practical guidance I can give you based on my experiences with these latest Googley gizmos and my intimate knowledge of their assorted Pixel predecessors. And the most important variable to consider is which specific Pixel you’re packin’ in your suspiciously sticky palm.
So think it through with me, won’t ya? And whether you end up snagging the latest and greatest Pixel or stickin’ with your current model, be sure to mosey your way over to my free Pixel Academy e-course next. It’ll teach you all sorts of useful new stuff your phone can do for you, no matter which specific Pixel you end up possessing.
Google Pixel 9 vs. Pixel 8 (or Pixel 8 Pro)
We’ll start with the most recent generation of Pixel devices — the same generation I’ve personally got in my oversized person-pocket this very minute.
The Pixel 9 Pro (left) with the Pixel 8 Pro (right) — different styles but similar sizes.
JR Raphael, IDG
(With this and all the subsequent comparisons, by the way, the same general advice applies regardless of whether you’re thinking about the Pixel 9, the Pixel 9 Pro, or the Pixel 9 Pro XL. This year, all of those devices are fairly similar and comparable in terms of their core qualities and capabilities — with the main differences being that the Pro-level Pixel 9 models have better displays, the added presence of a telephoto camera for extra-exceptional zooming (along with the new Zoom Enhance feature for extra-powerful post-capture cropping and zooming), and a better front-facing camera for shameless selfies. And, of course, the Pixel 9 Pro XL is bigger.)
This one’s a somewhat two-pronged answer: Whether you’ve got the Pixel 8 Pro or its smaller Pixel 8 sibling, your not-even-quite-one-year-old Pixel is still perfectly peachy — and still set to get Android updates for another solid six years yet. You’ve got no particularly pressing reason to make the leap to the Pixel 9, and you’d certainly be fine to stick with your current phone for another year or two at a bare minimum, maybe even longer.
That being said, as someone with a Pixel-8-generation phone, there’s a decent chance you’re a tech enthusiast and the type of person who simply likes having the best available product. So if you’re looking for a reason to upgrade, I’ll give you a handful (with advance apologies to your wallet):
Hardware design: It’s not a make-or-break update or anything that’ll have a massive impact on your day-to-day life, but the Pixel 9 devices really do have a sleek new look ‘n’ feel to ’em that’s quite enticing. It makes the previous-gen models feel a bit dated in comparison, once you get used to it. Google also says the new body makes the Pixel 9 twice as durable as its predecessors — a claim that’s tough to validate in any scientific way, but take it for what it’s worth.
Size without compromise: If you’re someone who prefers a smaller device but also wants the best of the best, the Pixel 9 marks the first time in a long time that you’ve got the ability to go compact without making any other sacrifices — via the smaller but otherwise equal Pixel 9 Pro (non-XL version).
Satellite backup: The Pixel 9 series includes two years of free access to Google’s new emergency satellite system, which lets you contact emergency services and share your location even if you’re in an area without regular cellular service. You probably (and hopefully!) won’t take advantage of that often, if ever, but it’s certainly a nice bit of added assurance to have.
Pixel Screenshots: The new and presently Pixel-9-exclusive app encourages you to capture screenshots whenever you see something you want to remember and then makes it easy to search through, ask questions about, and generally revisit that info later. It’s one of the few genuinely clever and practical applications of generative AI right now, and it really is a useful addition. (It’s possible that the app could come to older Pixel devices eventually, though Google has yet to confirm any specific plans around that.)
Call Notes: The other genuinely useful AI-powered addition on the Pixel 9 series is Google’s smart system for recording voice calls and serving up text transcripts and summaries right after. While you can emulate Call Notes on any Android device, it’s inevitably gonna be far less of a smooth, simple, and seamless setup. (Again, it’s possible this feature may make its way to older Pixel models at some point, but we don’t know for sure if or when that might happen yet.)
The Pixel 9 Call Notes feature (left) and Pixel Screenshots system (right) are productivity-aiding highlights of Google’s new Pixel devices.
JR Raphael, IDG
One big reason not to factor into your Pixel upgrading decision, if you ask me, is the thing that Google is emphasizing most prominently about the new Pixel 9 models — and that’s AI, in a more general sense, and the prominent new role played by the company’s next-gen Gemini virtual assistant on these devices.
The highly touted Gemini Live conversational chatbot experience might be the most effective illustration of all. This overview of the Gemini Live experience pretty much sums it up: It’s awkward, factually challenged, and very much a solution in search of a problem. You might enjoy playing around with it for a few minutes, but odds are, you won’t be pulling it up and conversing with it often — not for any productivity-related purposes, at least. (And remember, too: The Gemini Live feature is available only with a paid Gemini Advanced subscription, which you get for a year with any Pixel 9 Pro purchase but otherwise costs 20 bucks a month to continue using.)
And hey, who knows? Maybe this’ll evolve and improve over time. But you shouldn’t buy a piece of technology based on its theoretical future potential. You should buy it based on what it can do today. And today, there’s no sugarcoating it: The Gemini AI experience just isn’t great.
(The good news is that while the Pixel 9 does still signify the beginning of the end of Google Assistant as an Android phone and tablet entity, you can still opt to switch back to Google Assistant on all the Pixel 9 models as of now, if you’re ever so inclined.)
To be clear: These really are phenomenal phones — absolutely the best Pixels I’ve ever used and arguably among the best all-around smartphones created to date. But they’re those things in spite of all the overhyped AI hullabaloo, not because of it.
That aside, the Pixel 9’s camera setup is a nice little bump up from what the Pixel 8 series offered, and the new Add Me feature for group photos is a nifty trick and something you might appreciate once in a while. The Pro models also have a much better selfie camera than what was present in last year’s phones. If you take a lot of selfies and care about their quality, this might be significant for you. By and large, though, unless you’re the sort of person who stares intently at side-by-side images to search for subtle differences, I doubt you’ll notice much of a meaningful change in the types of non-selfie photos you get from these phones compared to the Pixel 8 models.
Now, don’t get me wrong: The Pixel 9’s camera capabilities are incredibly impressive! But so is the camera setup on the Pixel 8. For most common purposes and in terms of real-world impact, the difference is relatively subtle from one generation to the next. It’s a welcome bump forward, in other words, but we were already starting at such a high level that the realistic gains here aren’t exactly gonna be life-changing for most of us.
The same goes for other internal upgrades — like a 35% brighter display and 20% longer battery life, according to Google’s estimations. I don’t doubt that these claims are accurate, and they represent nice little boosts that any Pixel owner would eagerly accept. I just think you’d be hard-pressed to really be aware of ’em in everyday use, and I wouldn’t treat ’em as significant factors in any upgrade decision.
➜ The verdict, in short: If you’re itchin’ for something new and you’re the type of person who geeks out over the finer points of mobile technology (hiya!), you’ll be thrilled with the Pixel 9 upgrade coming from a Pixel-8-series phone. But at the same time, it’s certainly not a critical or particularly monumental upgrade, practically speaking, and there’s nothing in the new phone that you need or that’ll make a massive difference in your day-to-day life. The purely sensible advice would be to hang onto your current phone for a while longer — though particularly with the current trade-in values for Pixel-8-series phones, you can give yourself plenty of excuses for making the leap now, if you want. And as a fellow Pixel 8 series owner myself, I don’t think you’d regret it one bit if you do.
Google Pixel 9 vs. Pixel 8a
Got Google’s most recent midrange Pixel model? Moving to the Pixel 9 or Pixel 9 Pro would absolutely be an improvement for you, but that’s in large part because you’re moving from a more budget-minded proposition into premium flagship terrain.
The regular Pixel 9 (left) may be comparable in size to the Pixel 8a (right), but beyond the surface, it’s a whole different story.
JR Raphael, IDG
That means you’ll get a fancier-feeling and more premium phone body (for whatever that’s worth to you), along with a higher-quality screen and plenty of extra bells and whistles — including a slew of new camera-related goodies.
The Pixel 9 and Pixel 9 Pro are also in a completely different league when it comes to processing power, which can make the phone feel a fair bit snappier. That being said, the Pixel 8a is certainly no slouch with performance for most average-mammal needs, and you may or may not notice any obvious difference in your day-to-day use (especially if you aren’t doing a ton of multitasking and other high-resource-requiring work).
The biggest question to ask yourself is why you went with the Pixel 8a in the first place. If it’s because of the price and value, then moving up to the Pixel 9 or Pixel 9 Pro probably isn’t an advisable move. Your 8a still has a solid six and a half years (!) of active software support under its belt, and nothing about this upgrade would dramatically revolutionize your core user experience or what your phone’s capable of doing — outside of photography, at least, but even there, we’re talking more niceties and added abilities as opposed to core competency.
If past trends hold true, meanwhile, we’ll see the next midrange Pixel — the Pixel 9a — sometime next spring. So that’s also on the way as a possible upgrade within your current “a”-series path, should you want to move to something new but stay within the same pricing range.
➜ The verdict, in short: If you’re jonesing for a more premium Pixel experience, you’ll be delighted by the move to a Pixel 9 or Pixel 9 Pro — but that’s mostly because you’re hopping from one product tier to another. For most people who bought the 8a in the first place, sticking with that phone for another year or two, at a minimum, will make more sense.
Google Pixel 9 vs. Pixel 7 (or Pixel 7 Pro)
Still rockin’ one of 2022’s Pixel 7 series devices? You fall into a fun gray area of whether an upgrade to a Pixel 9 model would be worth your while.
The Pixel 9 Pro (left) is about the same size as the Pixel 7 Pro (right), though with a much rounder, sleeker-seeming body.
JR Raphael, IDG
As the last Pixel before Google moved from a three-year to its current seven-year software update promise, your device is only slated to receive Android operating systems updates through next fall — October of 2025. That means you could easily wait another year (or arguably even two) to make the leap to a new phone, and I don’t think you’d be missing out on anything incredibly important if you did.
That being said, you’d absolutely appreciate the same series of Pixel-9-introduced improvements we mentioned a moment ago, when talking about Pixel 8 owners. In addition to that, you’d gain the ability to use your phone’s Face Unlock system everywhere — as a secure way to sign into banking apps or anywhere else authentication is required — which is something that isn’t possible in the Pixel 7 series.
That improvement alone was what convinced me to pull the trigger on a Pixel 8 last year, when that capability first came into the equation. Small of a detail as it may seem compared to the more attention-grabbing Pixel tricks, it’s a real game-changer and something that has a meaningful, ongoing impact on your day-to-day life — likely multiple times every single day.
You’ll also gain the unusual advantage of an on-device thermometer, which seemed silly when it first debuted with the Pixel 8 series last year but has since grown into a valuable addition — now that it supports actual human temperature-taking. We can never find regular thermometers in my house, and even when we do, they’re frustratingly inconsistent and inaccurate in their readings. The Pixel 9’s thermometer (on the Pro models only) is always there and ready, and it’s been incredibly consistent with its measurements in my experience. (As luck would have it, an under-the-weather kiddo has given me plenty of opportunities to test it out.)
Still, this one mostly comes down to a matter of added niceties and conveniences — of want over need. If you want it, it’s well worthwhile, and you’ll certainly see enough of a difference to justify the change. But you definitely don’t need it, and if you aren’t feeling the urge to get something new, there’s really no reason not to stick with what you’ve got for another year or so.
➜ The verdict, in short: The Pixel 9 or Pixel 9 Pro would be a nice upgrade for you but by no means a necessary one. Ask yourself how much you truly want one of the devices and how much you can stomach the cost — taking into account any trade-in variables — for what’d be an inessential, if also enjoyable and efficiency-enhancing, purchase.
Google Pixel 9 vs. Pixel 7a
The story with the Pixel 7a is honestly pretty similar to what we said a moment ago with the Pixel 8a — though with the added consideration that the 7a will only receive Android OS updates through May of 2026.
That means, in all likelihood, that it’ll get this year’s upcoming Android 15 update, get next year’s Android 16 update, and then reach the end of its road in terms of ongoing OS support. That’s a decent amount of life left, if you aren’t in any rush to jump into something new. And even when you do replace your current phone, you may prefer to stay on the midrange path and go with the more affordable Pixel 8a (or a future Pixel “a” model) instead.
The Pixel 9 or Pixel 9 Pro would be a significant upgrade from the 7a in every possible measure, and you’d absolutely notice the difference. But your current phone is still fine. And just like with the Pixel 8a, the real question here is how badly you want to move to a more premium, high-end phone experience and how much you’re willing to spend to get it.
➜ The verdict, in short: For most 7a owners, there’s no pressing reason to upgrade — though going from that phone to the Pixel 9 or Pixel 9 Pro would be a substantial and immediately noticeable improvement, if you simply wanted to give yourself the added advantages of the more premium Pixel path without any more waiting.
Google Pixel 9 vs. Pixel 6 (or Pixel 6 Pro)
Pixel 6 pals, you’re next! And if you’ve been reading everything up through this point, you’ve probably got a pretty good guess of where we’re going with this.
With each subsequent generation of Pixel, the argument for making the upgrade gets more compelling — and that’s absolutely true in the case of the Pixel 6.
The Pixel 6 is set to receive current OS updates only through this coming October, which means you’ll receive Android 15 soon but then likely not see any additional OS updates after. That puts a firm, but not quite imminent, shelf life on your device, and you’d be well-advised to consider a new phone by sometime next year — ahead of Android 16’s release — at the very latest.
As for right now, upgrading to a Pixel 9 model would give you some noticeable, meaningful improvements — including the presence of Face Unlock for authentication (a huge quality-of-life upgrade, if you ask me!) along with a much snappier and more consistent fingerprint sensor as a backup option and some fairly significant camera upgrades. And, of course, you’d gain access to all the other productivity-boosting Pixel-9-specific elements we mentioned in the first section of this story.
➜ The verdict, in short: You could hold off one more year, if you’d rather, but upgrading to the Pixel 9 now would be a substantial upgrade for you — and the time is nigh for when such a move will become highly advisable. If you’re ready now, you’ll be delighted with the change the Pixel 9 brings you. If not, you could put the purchase off one more year and go with the Pixel 10 in 2025 instead.
Google Pixel 9 vs. Pixel 6a
Once again, the Pixel 6a situation is pretty similar to what we’ve described with other Pixel “a” models — though here, we’re looking at a two-year-old device down to its final year of active OS update support.
Practically speaking, the 6a’s OS update support end-date of July 2025 means that it’ll likely receive this year’s upcoming Android 15 update and then reach the end of the road for any additional OS updates beyond that. Even with security updates still in the picture, it wouldn’t be fully optimal and advisable to continue using past that point.
Still, after Android 15, we likely won’t see another major Android OS update until sometime next fall — and by then, there’ll be a whole new crop of Pixel products to choose from. So you could easily hold off a little while longer and then think about getting either a new midrange model or one of next year’s Pixel 10 phones as a replacement.
Once more, the Pixel 9 or Pixel 9 Pro would be a significant upgrade in every possible measure, and you’d absolutely notice and appreciate the difference. But, again, your current phone is still reasonably okay — if a little long in the tooth.
➜ The verdict, in short: The argument for upgrading from a Pixel 6a is a strong one, given how much things have moved forward in the time since that phone’s release — but still, this is far from a must-upgrade scenario. And, just like with the Pixel 8a, the real question is if you even want to move to the higher-end Pixel path whenever you do upgrade or if you’d rather stick with the less expensive midrange “a” phones. That’s a decision only you can make.
Google Pixel 9 vs. Pixel 5 or any earlier Pixel model
This last Pixel 9 upgrade decision is the easiest: If you’re using the Pixel 5 or any Pixel model earlier than that, you should strongly consider stepping up to the Pixel 9 or Pixel 9 Pro now.
These older Pixels are no longer actively being supported with current OS updates or even security patches, and all current features and added conveniences aside, that means they’re no longer advisable to use when it comes to the ever-important areas of optimal privacy, security, and performance for your phone.
➜ The verdict, in short: It’s time for a new device, plain and simple, and the Pixel 9 models will be a tremendous, night-and-day leap forward for you in every possible measure. If you’re on an older Pixel “a” model, you could consider going with the Pixel 8a instead, should you wish to spend a little less dough. But with any older Pixel, the Pixel 9 or Pixel 9 Pro would be the best upgrade you could make right now, and you would be positively thrilled with the change (in addition to being maximally protected, in terms of being on the most current software possible and avoiding any privacy-, security-, or performance-related liabilities).
Folding phones are their own beast, and it’s a bit of an apples-to-oranges comparison to try to stack them up next to a standard slab-style device at this point.
If you simply aren’t happy with your Fold experience and think you’d rather go back to a more standard sort of phone form, you’d be quite pleased with a move to a Pixel 9 model.
Otherwise, stay tuned for more on the second-gen Pixel Fold — a.k.a. the Pixel 9 Pro Fold — soon. It isn’t slated to ship until sometime in September, and Google consequently hasn’t made review units of it available for assessment quite yet.
Don’t let yourself miss an ounce of Pixel magic. Come start my free Pixel Academy e-course to discover tons of fresh hidden features and time-saving tricks for whatever Pixel phone you’re using.
From the editors of Computerworld, this enterprise buyer’s guide helps IT staff understand what the various remote IT support tools can do for their organizations and how to choose the right solution.
Microsoft announced on Wednesday that it will begin testing its controversial “Recall” AI search and recall feature for Windows Insiders in October. Earlier, it was slated for launch in June.
“As previously shared on June 13, we have adjusted our release approach to leverage the valuable expertise of our Windows Insider community prior to making Recall available for all Copilot+ PCs,” Microsoft said in a blog post. “With a commitment to delivering a trustworthy and secure Recall (preview) experience on Copilot+ PCs for customers, we’re sharing an update that Recall will be available to Windows Insiders starting in October.”
The Recall feature captures screenshots of on-screen activity, allowing users to search for information they saw or searched previously.
However, it immediately raised concerns among security researchers that automatically capturing images without explicit user consent violates user privacy and could make sensitive personal information more accessible to attackers.
In response to these concerns, Microsoft stated in June that the Recall feature would be disabled by default and pledged to implement additional security enhancements.
“We are adjusting the release model for Recall to leverage the expertise of the Windows Insider community to ensure the experience meets our high standards for quality and security,” the software major said in June.
“For features such as Recall, ideally the data should be stored and processed completely ‘on device’ locally and data shouldn’t leave the laptop,” said Neil Shah, VP for research and partner at Counterpoint Research. “This will drive the real on-device, privacy-centric AI promise. If the model has to learn from user’s data and habits, it should also reside locally with the flexibility to encrypt the data and the model on-device.”
Microsoft has not specified a timeline for a broader release of Recall to all Windows PCs that meet the system requirements for Copilot+ PCs.
Copilot+ PCs are a new class of Windows devices from various manufacturers that can run AI workloads. Microsoft unveiled Recall running on these devices at an event in May.
The timing of the Recall feature’s wider release could be critical, particularly with the upcoming holiday season. Consumers may be more inclined to purchase new devices if Recall is made available across all compatible PCs by then.
Why is Recall a concern?
Windows Recall is a new feature that is designed to come with new Copilot+ PCs, which Microsoft announced in May. This AI-powered tool takes screenshots of your screen every five seconds allowing you to search through a log of your past activities for up to three months.
The screenshots are stored and processed on your device, secured with encryption. You have the option to exclude specific apps and websites from being recorded, and you can pause the Recall feature whenever needed.
The concerns arise from two aspects. First, it is “turned on” by default, as per the initial announcement, and can record and store user data without obtaining explicit consent. Second, it does not conceal or hide sensitive data including passwords or financial data, that might appear on your screen.
Device makers are keen to demonstrate that users can run AI models on their local PCs, bypassing the need for cloud-based services from companies like OpenAI. Apple has similarly equipped its latest MacBooks with the ability to run AI models locally.
“However, some generative AI-centric features for CoPilot will require cloud-based processing for tasks like information retrieval, search, or querying,” Shah said. “Ensuring that data remains secure—whether on the device, in transit, or in the cloud—will be a key challenge. This aspect will also be a critical focus and differentiator for companies like Microsoft compared to Apple in the PC space.”
Security has become an increasing priority for Microsoft, especially after a Department of Homeland Security report in April raised concerns about China’s breach of US government officials’ Microsoft-based email accounts.
As Microsoft moves forward with the testing and potential rollout of Recall, the company will need to balance innovation with user privacy and security concerns, especially as it positions itself in the competitive AI and PC markets.
Microsoft announced on Wednesday that it will begin testing its controversial “Recall” AI search and recall feature for Windows Insiders in October. Earlier, it was slated for launch in June.
“As previously shared on June 13, we have adjusted our release approach to leverage the valuable expertise of our Windows Insider community prior to making Recall available for all Copilot+ PCs,” Microsoft said in a blog post. “With a commitment to delivering a trustworthy and secure Recall (preview) experience on Copilot+ PCs for customers, we’re sharing an update that Recall will be available to Windows Insiders starting in October.”
The Recall feature captures screenshots of on-screen activity, allowing users to search for information they saw or searched previously.
However, it immediately raised concerns among security researchers that automatically capturing images without explicit user consent violates user privacy and could make sensitive personal information more accessible to attackers.
In response to these concerns, Microsoft stated in June that the Recall feature would be disabled by default and pledged to implement additional security enhancements.
“We are adjusting the release model for Recall to leverage the expertise of the Windows Insider community to ensure the experience meets our high standards for quality and security,” the software major said in June.
“For features such as Recall, ideally the data should be stored and processed completely ‘on device’ locally and data shouldn’t leave the laptop,” said Neil Shah, VP for research and partner at Counterpoint Research. “This will drive the real on-device, privacy-centric AI promise. If the model has to learn from user’s data and habits, it should also reside locally with the flexibility to encrypt the data and the model on-device.”
Microsoft has not specified a timeline for a broader release of Recall to all Windows PCs that meet the system requirements for Copilot+ PCs.
Copilot+ PCs are a new class of Windows devices from various manufacturers that can run AI workloads. Microsoft unveiled Recall running on these devices at an event in May.
The timing of the Recall feature’s wider release could be critical, particularly with the upcoming holiday season. Consumers may be more inclined to purchase new devices if Recall is made available across all compatible PCs by then.
Why is Recall a concern?
Windows Recall is a new feature that is designed to come with new Copilot+ PCs, which Microsoft announced in May. This AI-powered tool takes screenshots of your screen every five seconds allowing you to search through a log of your past activities for up to three months.
The screenshots are stored and processed on your device, secured with encryption. You have the option to exclude specific apps and websites from being recorded, and you can pause the Recall feature whenever needed.
The concerns arise from two aspects. First, it is “turned on” by default, as per the initial announcement, and can record and store user data without obtaining explicit consent. Second, it does not conceal or hide sensitive data including passwords or financial data, that might appear on your screen.
Device makers are keen to demonstrate that users can run AI models on their local PCs, bypassing the need for cloud-based services from companies like OpenAI. Apple has similarly equipped its latest MacBooks with the ability to run AI models locally.
“However, some generative AI-centric features for CoPilot will require cloud-based processing for tasks like information retrieval, search, or querying,” Shah said. “Ensuring that data remains secure—whether on the device, in transit, or in the cloud—will be a key challenge. This aspect will also be a critical focus and differentiator for companies like Microsoft compared to Apple in the PC space.”
Security has become an increasing priority for Microsoft, especially after a Department of Homeland Security report in April raised concerns about China’s breach of US government officials’ Microsoft-based email accounts.
As Microsoft moves forward with the testing and potential rollout of Recall, the company will need to balance innovation with user privacy and security concerns, especially as it positions itself in the competitive AI and PC markets.
Market research firm Gartner yesterday published its 2024 Hype Cycle for Emerging Technologies, and the study revealed that generative AI (genAI) has passed the “peak of inflated expectations” and is now sliding down into the “trough of disillusionment.”
Along with genAI, AI-augmented software engineering is also heading down the slope, after passing its inflated expectations in markets, according to Gartner, whose Hype Cycle describes the hot ascent and eventual cooling off of technology adoption.
AI-assisted code generation tools are increasingly prevalent in software engineering, and somewhat unexpectedly have become low-hanging fruit for most organizations experimenting with genAI. Adoption rates are skyrocketing. That’s because even if they only suggest a baseline of code for a new application, automation tools can eliminate hours that otherwise would have been devoted to manual code creation and updating.
Hitting the peak of inflated expectations is prompt engineering, according to Gartner. While most large language models like OpenAI’s GPT-4 are pre-filled with massive amounts of information, “prompt engineering,” a way of training the algorithm, allows genAI to be tailored for specific industry or even organizational use.
GenAI interest wanes as ROI becomes the focus
Excitement around foundation models, such as Google Gemini, Anthropic Claude, Amazon Bedrock, and OpenAI GPT-4, is waning among enterprises as companies instead seek concrete returns on investment (ROI). These days, companies are more often than not deploying genAI only for use cases that drive ROI, according to Arun Chandrasekaran, a Gartner distinguished vice president analyst.
“Generative AI is sliding through the trough of disillusionment due to mismatch between high expectations vs. reality, enterprise challenges in maturing their data engineering and AI governance, as well as intangible ROI of many genAI initiatives,” Chandrasekaran said.
While the technology has been heralded as a boon to productivity, nailing down an ROI in genAI can prove to be elusive. That’s not necessarily because finding ROI is difficult, but expressing ROI has been difficult because many benefits like productivity have indirect or non-financial impacts that create financial outcomes in the future, according to Rita Sallam, a distinguished vice president analyst at Gartner.
Gartner’s trough of disillusionment describes a time when interest wanes as experiments and implementations fail to deliver on the initial hype of a technology. Producers of the technology shake out or fail. Investment continues only if the surviving providers improve their products to the satisfaction of early adopters, according to Gartner.
AI agents step into the spotlight
But far from a negative effect, the trough of disillusionment can lead to what Gartner describes as the “plateau of productivity,” when mainstream adoption starts to take off. It also means enterprise focus on ROI will likely spur adoption of autonomous AI in the form of AI agents — something with a more solid potential for productivity and efficiency gains.
An AI agent is a software program that collects data and uses the data to perform self-determined tasks to meet predetermined goals. For example, an AI agent could act as a customer care representative and automatically ask the customer different questions, look up information in internal documents, and respond with a solution. Based on the customer responses, it determines if it can resolve the query itself or pass it on to a human.
By 2030, companies will spend $42 billion a year on genAI projects such as chatbots, research, writing, and summarization tools, according to Gartner.
Autonomous AI systems can operate with minimal human oversight. They seek to “understand” their environment, draw conclusions from it and adjust their actions accordingly, according to Chandrasekaran.
“They can make decisions, purchase things and perform tasks, achieving goals in a range of environments as effectively as humans can. Systems that can perform any task a human can perform are beginning to move slowly from science fiction to reality,” he said.
While the current generation of AI models lack “agency,” AI research labs are quickly releasing agents that can dynamically interact with their environment to achieve goals, although it will be a gradual process, Chandrasekaran noted.
An eye on other emerging tech
“Even as AI continues to grab the attention, CIOs and other IT executives must also examine other emerging technologies with transformational potential for developers, security, and customer and employee experience and strategize how to exploit these technologies in line with their organizations’ ability to handle unproven technologies,” Chandrasekaran said.
Gartner said its Hype Cycle for Emerging Technologies is unique among the company’s other Hype Cycles because it distills insights from more than 2,000 technologies and focuses on “must-know” emerging technologies.
“These technologies have potential to deliver transformational benefits over the next two to 10 years,” Gartner said.
Autonomous AI software was among four emerging technologies called out in the report because it can operate with minimal human oversight, improve itself, and become effective at decision-making in complex environments.
“These advanced AI systems that can perform any task a human can perform are beginning to move slowly from science fiction to reality,” Gartner said in its report. “These technologies include multiagent systems, large action models, machine customers, humanoid working robots, autonomous agents, and reinforcement learning.”
Autonomous agents are currently heading up the slope to the peak of inflated expectations. Just ahead of autonomous agents on that slope is artificial general intelligence, currently a hypothetical form of AI where a machine learns and thinks like a human does.
GenAI technologies are evolving at a rapid pace, Chandrasekaran noted, and the innovation continues at a rapid pace, which can be overwhelming enterprise IT leaders.
“Many enterprises are also realizing that genAI alone may not be a panacea for all their use cases, and they need to combine it with other AI techniques for meaningful value,” Chandrasekaran said. “The long-term potential of generative AI will still be significant, but enterprise IT leaders need to address the near-term risks to reach the plateau of productivity.”
OpenAI has expressed opposition to California’s proposed bill regulating AI development and deployment as lawmakers prepare for an upcoming vote.
In a letter addressed to California State Senator Scott Wiener, the AI startup argued that the bill would stifle innovation in the sector and suggested that such regulation should be handled at the federal rather than state level, according to a Bloomberg report.
But in response, Wiener said that the OpenAI letter doesn’t criticize a single provision of the bill and that the company appears to acknowledge the bill’s specific core provisions.
“Instead of criticizing what the bill actually does, OpenAI argues this issue should be left to Congress,” Wiener said in a statement. “As I’ve stated repeatedly, I agree that ideally, Congress would handle this. However, Congress has not done so, and we are skeptical Congress will do so.”
“Under OpenAI’s argument about Congress, California never would have passed its data privacy law, and given Congress’s lack of action, Californians would have no protection whatsoever for their data,” Wiener added.
The bill, SB 1047, seeks to implement safety regulations for large-scale AI models that surpass certain size and cost benchmarks. Passed by the state Senate in May, the legislation mandates AI firms to adopt measures ensuring their technologies do not facilitate severe risks, including the creation of bioweapons capable of widespread casualties or causing financial losses exceeding $500 million.
A controversial bill
Other technology firms, including Meta and Alphabet, and trade associations including the AI Alliance have also reportedly opposed the bill.
Charlie Dai, VP and principal analyst at Forrester, noted that although AI governance on security, privacy, and regulatory compliance is crucial, the new bill could create unnecessary business uncertainty and raise operational costs for most AI companies.
This may slow the pace of innovation and harm the overall open source ecosystem surrounding AI.
“AI firms need to consider a range of options to mitigate the effects, such as engaging with lawmakers collaboratively to shape the bill in a way that balances safety with innovation, making more investment in regulatory compliance, and expanding or relocating to states with more flexible policies,” Dai added.
In the letter, OpenAI warned that the proposed legislation could significantly and adversely impact US competitiveness in AI and national security.
Wiener responded that, far from undermining national security, SB 1047’s requirements for AI companies to thoroughly test their products for the ability to cause catastrophic harm can only strengthen national security.
Not limited to California-based companies
OpenAI has also argued that the legislation could drive companies out of California, but Senator Wiener countered, noting the bill would affect any company doing business in the state, regardless of where they are based.
“This tired argument — which the tech industry also made when California passed its data privacy law, with that fear never materializing — makes no sense given that SB 1047 is not limited to companies headquartered in California,” Wiener said in the statement. “Rather, the bill applies to companies doing business in California. As a result, locating outside of California does not avoid compliance with the bill.”
Efforts have been made throughout the year in collaboration with open-source advocates, Anthropic, and others to refine and improve the bill, according to Wiener. He added that SB 1047 is well-calibrated to address foreseeable AI risks and deserves to be enacted.
OpenAI has expressed opposition to California’s proposed bill regulating AI development and deployment as lawmakers prepare for an upcoming vote.
In a letter addressed to California State Senator Scott Wiener, the AI startup argued that the bill would stifle innovation in the sector and suggested that such regulation should be handled at the federal rather than state level, according to a Bloomberg report.
But in response, Wiener said that the OpenAI letter doesn’t criticize a single provision of the bill and that the company appears to acknowledge the bill’s specific core provisions.
“Instead of criticizing what the bill actually does, OpenAI argues this issue should be left to Congress,” Wiener said in a statement. “As I’ve stated repeatedly, I agree that ideally, Congress would handle this. However, Congress has not done so, and we are skeptical Congress will do so.”
“Under OpenAI’s argument about Congress, California never would have passed its data privacy law, and given Congress’s lack of action, Californians would have no protection whatsoever for their data,” Wiener added.
The bill, SB 1047, seeks to implement safety regulations for large-scale AI models that surpass certain size and cost benchmarks. Passed by the state Senate in May, the legislation mandates AI firms to adopt measures ensuring their technologies do not facilitate severe risks, including the creation of bioweapons capable of widespread casualties or causing financial losses exceeding $500 million.
A controversial bill
Other technology firms, including Meta and Alphabet, and trade associations including the AI Alliance have also reportedly opposed the bill.
Charlie Dai, VP and principal analyst at Forrester, noted that although AI governance on security, privacy, and regulatory compliance is crucial, the new bill could create unnecessary business uncertainty and raise operational costs for most AI companies.
This may slow the pace of innovation and harm the overall open source ecosystem surrounding AI.
“AI firms need to consider a range of options to mitigate the effects, such as engaging with lawmakers collaboratively to shape the bill in a way that balances safety with innovation, making more investment in regulatory compliance, and expanding or relocating to states with more flexible policies,” Dai added.
In the letter, OpenAI warned that the proposed legislation could significantly and adversely impact US competitiveness in AI and national security.
Wiener responded that, far from undermining national security, SB 1047’s requirements for AI companies to thoroughly test their products for the ability to cause catastrophic harm can only strengthen national security.
Not limited to California-based companies
OpenAI has also argued that the legislation could drive companies out of California, but Senator Wiener countered, noting the bill would affect any company doing business in the state, regardless of where they are based.
“This tired argument — which the tech industry also made when California passed its data privacy law, with that fear never materializing — makes no sense given that SB 1047 is not limited to companies headquartered in California,” Wiener said in the statement. “Rather, the bill applies to companies doing business in California. As a result, locating outside of California does not avoid compliance with the bill.”
Efforts have been made throughout the year in collaboration with open-source advocates, Anthropic, and others to refine and improve the bill, according to Wiener. He added that SB 1047 is well-calibrated to address foreseeable AI risks and deserves to be enacted.
Apple’s iPhone and App Store turned the mobile phone industry upside down, created the smartphone generation, and set the scene for developer success that did not exist before, all while working to protect privacy and security.
No wonder Apple’s enemies want to break all that the company has achieved. No wonder they hope to feast on the crumbs left behind. The regulators seem to want to let them do just that, but what choice will consumers be given as they endure the tyranny of choice?
But, as the company has argued, some of these moves can, may, or will erode platform security, which is something many of its customers expect from its products. Surely those customers deserve to keep to that choice, too?
However, the regulators don’t seem to see it that way, insisting on changes to Apple’s iPhone platform that, quite frankly, threaten to turn it into the kind of flimsy, compromised beast we might have had if Windows had won the mobile war.
Luckily, Windows failed to win that war.
That’s not to say that all the arguments to force Apple to open up are flimsy. Apple does have huge market power, it can enter new markets fairly easily, and it seems appropriate to find ways to create new opportunities across its platforms.
But should those opportunities replace the existing privacy and security Apple’s customers luxuriate in today? Surely that privacy and security is also a choice.
Privacy and security should be an option
Perhaps there is a way Apple can provide both things: the essential curated experience hundreds of millions of us already love, and the more open platform its competitors seek to draw profits from. Perhaps it’s time to fork the platform.
Think about it this way — it seems the introduction of support for third-party app stores and so on is being forced on Apple as a universal constraint. But should it be? Shouldn’t Apple’s customers have the right to choose which way to go?
Many may decide to work with third-party app stores so they can use alternative billing systems and make that bloke from Epic Games even richer, but many others may never, ever want to play those games and may instead want to remain entirely in Apple’s so-called “walled garden.” Why shouldn’t they be able to?
What about giving people choice?
An interesting addition over the last 12 months on iOS has been a new and simpler way to run iOS beta software on your device. You can now do so with the flick of a switch.
What if Apple used that same system to deliver two breeds of its standard operating system? The first would be the iOS we all love and use today, though likely with the addition of new APIs to make some functions (such as mobile payments) more competitive; the second might be a more open version of iOS, equipped with support for external stores, payment systems, and all the other things people with lots of money seem to get angry about when it comes to Apple’s systems.
That’s a compromise, perhaps, but it means Apple’s customers could vote with their own fingers. They could choose to join life outside the garden or stay within it. That is, after all, a choice they should be able to make. For many users, it is the choice they already took when they selected Apple’s platforms.
Some people need security more than they need Fortnite
It’s also a choice many enterprise users of Apple products want to make.
Particularly in regulated industries, they need to ensure the privacy and security of sometimes highly sensitive data. To do so they need — they are actually legally required — to ensure every possible protection is in place.
Choosing Apple’s hard-as-nails walled garden iOS would be the option they took with their managed devices. People regularly accessing your medical data on a device shouldn’t be installing software that may or may not be completely safe from third-party stores that may or may not be what they seem. Many companies forbid the use of all kinds of device features using MDM controls, and taking the choice to remain all-in on Apple’s model is a choice they probably want to keep.
Maybe there’s another way
Perhaps Apple is thinking in the same way, particularly following the shock resignation of Matt Fischer, Apple’s Worldwide head of the App Store and the decision to split those operations into two segments: one to handle the App Store as is, the other to handle incoming alternative distribution systems. I don’t know if Apple is thinking in this direction; I’m merely speculating that it could be.
If it were, then it would provide a choice that lets people currently using iPhones retain the right to keep things as they are, rather than being forced to open up because a smattering of well-connected millionaires want to make money out of their insecurity. A lot of people — customers, developers and not just Apple — have already made a great deal of money while also protecting their security, after all.
If Apple moves in that direction, the usual chorus of voices, amplified by a click-bait-hungry media, will castigate the company for the new buzz word of “malicious compliance.” But the question, at least when it comes to customers happy with the status quo, is why should they be forced to accept an openness they neither want nor need?
