Computerworld.com [Hacking News]

For the past two or three weeks, upstart social network Bluesky has been attracting large numbers of new users every day, a crowd that quite evidently includes many Apple device users — the app has at times topped Apple’s App Store charts in the run-up to Thanksgiving in the US. I’ve been using it for a little while and have collected a handful of hints and services that can optimize the experience for anyone migrating to the increasingly busy social network.

What’s interesting about Bluesky is that while it lacks some of the features you can find on other social sites, the conversation seems subjectively better and the lack of ads and sponsored posts — along with the absence of algorithmic recommendations from people you don’t want to hear from — make for a pleasant social media experience. 

A place for friends?

One thing about the service is that it is a good internet citizen, which means it does not throttle outbound links and does not punish posters by reducing visibility of their posts simply for sharing them. 

Among other things, this means traffic from the service to The Boston Globe is already three times what that publication achieves from Threads. It also means the Globe is seeing significantly more of those readers converting to a subscription account.

In other words, while at roughly 22 million users, Bluesky remains a relatively small service. But the people using it appear to be actively doing so, which means it can generate decent traffic. That openness also means that data you share via the service is in the clear and can and will be picked up to train AI models and so on. There is no privacy there, and this should be clearly understood — everything you post is out in the open, so don’t share what you don’t want everyone to know. At the same time, by virtue of the service being so open, it can deliver an even better experience with the help from third-party tools and services like these.

Thread summaries

One thing we all used to use in X, thread summaries aren’t natively available on Bluesky, but you can use Skyview to quickly and easily create summaries. Just mention @skyview.social with “unroll” anywhere in a thread and receive a link to view/share the entire conversation. The downside might be that all the unroll requests end up being visible on Skyview’s account (though to be fair, everything you do is already stored somewhere on every social media service — it’s just that Skyview lets you see it, too).

Publish threads

Back in the olden days it was possible to publish entire conversations using a link from what was then Twitter. These turned up across numerous websites and within innumerable stories. Then it got switched off. The good news is that to some extent the good times are back. To turn a thread into a shareable web page use Skywriter.blue, which turns a Bluesky post URL into a shareable webpage. While this doesn’t yet work as an embed, it is at least now a page.

Mute a thread

You might need to know this in case you get involved in a conversation that blows up and you don’t want to receive additional alerts concerning it. To mute a thread, just tap the three dots on the right of any post in that thread and then choose “Mute Thread” from the options there. Additional tools you’ll find when you tap those three dots include:

• Translate
• Copy post text
• Send via direct message
• Share
• Mute words & tags
• Hide reply for me.

What lists are you on?

If you are interested in finding out what lists you are on, who you have blocked, and who has blocked you, visit clearsky.app, enter the relevant handle, and take a look.

Find more Starter Packs

You’ll find a directory of all Starter Packs — collections of recommended people to follow usually chosen to represent specific topics or subject areas — at Blueskydirectory.com. Explore here to find more collections from which to cherry-pick those you wish to follow, or just explore the comic genius of this Starter Pack about men called Geoff.

Starter Packs can become lists

BlueSky users are incredibly busy building Starter Packs. They cover multiple bases — independent book publishers, RStats Ecosystem Maintainers, dogs of Bluesky, for example — and are maintained and provided by volunteers. The idea is that you can visit these collections and select people you want to follow or follow the entire list. It’s a good way to build a solid selection of feeds for your interests.

Lists are another thing. Lists are groups of accounts that can be used to curate feeds, follow interests, and so on. You might use these to track your favorite writers, researchers, or other high-profile people you want to keep an eye on — that way you don’t need to follow them. The great thing about lists is that, unlike Starter Packs, you can choose to create a separate feed in your own window that follows posts from that list; I follow the BlueSky Team list, for example. This lets me monitor posts in that list without following more people.

But what if you want to turn a Starter Pack into a list, so you can create your own window to follow? You can! Just use the Pack2List web app, where you can paste the URL of that Starter Pack and choose to follow it as a list. That means that all the content in one of those Starter Packs can be made easily available on your account without you following each person on the list.

More people to follow

One quick way to find other like-minded souls might be to find people you do not yet follow who are followed by lots of people you do follow. You do this using the Bluesky network analyzer, which you’ll find right here. 

Use Bluesky like Tweetdeck

Anyone out there with a memory that extends further than a few of months might remember Tweetdeck, the incredibly useful tool used by so many professionals to read and manage Tweets. Something very like it is now available to Bluesky users. Deck.Blue is an app you can connect to an account that lets you explore your “skeets’ in a multi-column layout, so you can monitor all your posts, posts from lists you follow, searches, notifications, hashtags — whatever you need. You can also use Deck.Blue to schedule posts, though this is currently a fee-based feature. The tool also works with Buffer, so you can monitor your LinkedIn, Mastodon, and Bluesky channels in one place.

Search tools

The search tools in Bluesky can help you in several ways, for example:

• Find a specific phrase: Use quotes around keywords, such as “Austrian Airlines” to find skeets that include that phrase.
• Hashtag search: Use any term with an # prefix to find any post with that tag. 
• Find people: You can search for posts that mention specific people using their Bluesky handle or use from: and a user handle to find all posts from a user. 
• URL: Paste in a URL to see all the posts that have shared that URL.

There are many more. Two additional third-party tools you should find useful are Skythread, which lets you search for a thread and comments using the thread link, and Label Scanner, which enables you to verify which labels have been applied to an account.

Alternative clients

Bluesky offers its own client, but the experience on an iPad is pretty bad. If you want to use the service on an iPad, take a look at Skeets, which is better and includes a host of additional features, some of which you must pay for. Other alternative Bluesky browsers include the aforementioned Deck.blue, SkyFeed, and SkeetDeck. There are more, and with millions migrating to the service, it is likely additional options will appear pretty swiftly for a while.

You will need an app password to make any of these services work. Rather than using your actual account password, I utterly urge you to create an auxiliary app password. You can use this to give the clients the access they need to work for you, while keeping your primary password secure. Finally, for even more insights into using Bluesky, visit this page.

You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

Download the PDF Computerworld Cheat Sheet today.

Windows Recall is an AI-based feature that will be built into new Copilot+ PCs. Recall takes snapshots of your screen every few seconds and uses that data to make a searchable index of everything you’ve ever done on your PC. But, since that first controversial announcement earlier this year, Microsoft has delayed Recall and made some changes.

In late November 2024, Recall became available in a preview form for Windows Insiders running the right hardware. Now it’s out in the wild.

So is there cause for concern about Recall and your privacy — or the privacy of data connected to your company? I’ve gotten a lot of questions about that from readers. Let’s dig in so you can understand exactly what’s going on and what decisions you’ll have to make, whether you’re thinking about your work laptop, a home PC, or a fleet of business computers.

But first, the good news: Windows Recall is only available on those new Copilot+ PCs. It won’t arrive on your current Windows 11 or Windows 10 PC with an update. You won’t even have to think about Recall until you buy a new PC branded a “Copilot+ PC.”

Want expert insights on what’s actually going on in Windows? Sign up for my free Windows Intelligence newsletter — three things to try every Friday. Plus, get free Windows Field Guides as a bonus when you sign up!

What is Recall on Windows?

Recall will be built into Windows 11 moving forward. And, again: It will only be available on new PCs Microsoft has certified as “Copilot+ PCs.”

These PCs have neural processing units (NPUs) capable of at least 40 trillion operations per second (TOPS.) This hardware is designed to accelerate local AI tasks — in other words, AI tasks performed on your PC without any online servers involved — in a way that doesn’t drain a lot of battery power.

If you do have a Copilot+ PC, Recall is an optional feature that can capture snapshots of your screen every five seconds. (It won’t capture audio or video — just pictures of what’s on your screen.) You will then be able to search those snapshots using plain-language search. For example, you could say: “Show me that PowerPoint presentation I was looking at three weeks ago, the one with the green bar chart” — or “What was that message Dave sent me about the quarterly budget two months ago?” These searches all happen entirely on your device, and they’ll even work offline. Microsoft’s servers aren’t involved.

It’s a more plain-language way of your computer remembering what you were doing and allowing you to dig through it — it’s clear how this type of feature could boost productivity for anyone who works on their PC — and anyone using it for other tasks, from online shopping to vacation planning to chatting with friends.

Macs have something similar with Rewind, which also captures your computer activity and lets you search it. But Rewind is a third-party tool, not something built into macOS by Apple.

srcset="https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?quality=50&strip=all 2272w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=300%2C168&quality=50&strip=all 300w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=768%2C432&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=1024%2C576&quality=50&strip=all 1024w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=1536%2C864&quality=50&strip=all 1536w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=2048%2C1152&quality=50&strip=all 2048w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=1240%2C697&quality=50&strip=all 1240w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=150%2C84&quality=50&strip=all 150w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=854%2C480&quality=50&strip=all 854w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=640%2C360&quality=50&strip=all 640w, https://b2b-contenthub.com/wp-content/uploads/2024/05/1-Microsoft-Windows-Recall-privacy-Search.png?resize=444%2C250&quality=50&strip=all 444w" width="1024" height="576" sizes="(max-width: 1024px) 100vw, 1024px">Windows Recall doesn’t just take screenshots — it uses AI models to analyze the contents so you can search for them in plain language.

Microsoft

What’s changed since the original announcement?

Since the original announcement of Recall in May, Microsoft has repeatedly delayed it and announced changes to boost privacy and security and make the feature more trustworthy. Here, specifically, is what’s evolved:

• Recall will be off by default unless you choose to turn it on during the Windows setup process.
• Recall will filter out sensitive information like password, credit card details, and social security numbers by default.
• Recall requires you use Windows Hello authentication before you can access your snapshots.
• Recall’s data is securely encrypted in a VBS Enclave, which means other users and applications can’t access it. The key is only released when you authenticate with Windows Hello.
• Recall will be disabled by default on managed business PCs. Businesses will have to choose to enable it. If they don’t, their employees won’t be able to activate it.
• Recall will be tested with Windows Insiders before it arrives on stable Copilot+ PCs. (Testing began in late November 2024.)

These changes were detailed by Microsoft in a blog post in June 2024 as well as an update on Recall’s architecture released in September 2024.

Overall, it’s a reasonable selection of changes that addresses many of the criticisms people had with the way Recall was originally implemented. It’s also led to a slower release with more testing — it doesn’t feel like a frantic launch that’s catching people off guard, like the original announcement did.

Is Microsoft sneaking Recall onto my PC with an update?

No. I can’t state this emphatically enough: Recall will not arrive on your current Windows 11 or Windows 10 PC. As Microsoft puts it, this feature is “exclusive” to those new Copilot+ PCs. It won’t suddenly arrive on any of your existing PCs via a Windows Update or any other mechanism.

As of November 2024, Copilot+ PCs are laptops with Qualcomm Snapdragon X Elite or Snapdragon X Plus processors, Intel Core Ultra (Series 2) “Lunar Lake” processors, or AMD Ryzen AI 300 series processors.

Do I have to use Recall?

Recall is completely optional. When you’re signing into a new Copilot+ PC for the first time, Microsoft says you’ll be informed about Recall and allowed to decide. You can choose not to use Recall at that time, in which case it won’t do anything or collect any sort of data.

If you do enable it, you’ll see a Recall icon pinned to your taskbar by default, and Recall will have a system tray icon while it’s running. It’s very visible — it doesn’t just run silently in the background. After all, Microsoft wants you to use Recall to find things.

Microsoft says you can use the system tray icon or the options at Settings > Privacy & security > Recall & snapshots to pause Recall at any time. You can turn it on or off, delete existing snapshots, and choose to filter specific apps and websites so Recall doesn’t capture them. Recall also won’t capture any activity in “private browsing” windows in browsers like Chrome, Edge, and Firefox.

The Settings app provides a lot of options for controlling Recall’s snapshots — and deleting them.

Microsoft

How can businesses control Recall?

Microsoft says Recall will be disabled by default on managed business PCs. Organizations won’t have to do anything to disable Recall on their devices. In fact, they’ll have to go out of their way to activate it if they want their employees to have access.

Businesses that do want to activate Recall can use either group policy or MDM (mobile device management) policy. Microsoft has a guide to controlling Recall for IT administrators.

Where are the Recall snapshots stored?

Recall stores all the snapshots and other data on your PC. When you perform a search, Recall does the search on your PC. Microsoft says your data is never uploaded to a Microsoft server. It all happens completely locally, without the computer ever “phoning home.”

In a way, this makes Recall a little less useful — if you use multiple PCs, your Recall activity won’t sync between them. If you’re looking for something, you’ll need to search Recall on the PC you originally saw it on. But that may be a good thing when it comes to privacy considerations, particularly from an enterprise perspective.

The Recall data is also stored in an encrypted manner that’s specific to each individual user account on a device.

You won’t be able to access Recall to see any of that activity from that device without first authenticating with Windows Hello authentication. You’ll need your face, fingerprint, or PIN to activate it, so people sitting down in front of your PC won’t just be able to see your Recall data; it’s encrypted with a key that’s only released to make it accessible when that authentication happens.

Is Microsoft taking my Recall data?

Microsoft says that the Recall data will be stored only on your PC and never processed by its servers. Since Microsoft isn’t so much as ever seeing or receiving this data, your Recall snapshots won’t be used for targeting ads to you, training AI models, or any other purpose along those lines.

Couldn’t someone steal my laptop and look at my snapshots?

Modern Windows PCs have encrypted storage, like other modern devices. Someone who stole your PC would need to be able to sign in as you to see your data.

Recall is only going to be available on Copilot+ PCs, and Microsoft has set a higher baseline of security for these PCs: They must be Secured-core PCs, for example, and they will include a Microsoft Pluton security processor. In other words, they will ship with encrypted secure storage backed by hardware security features.

The reality is that if someone stole a PC from an office worker or a home PC user and managed to sign into it, they’d already have access to a lot of private data. This would include financial documents stored on the PC itself, sensitive business information, email accounts the computer was signed into, and so on.

Recall will definitely generate extra data that can be accessed if a criminal breaks into a PC. But, on the whole, it’s less risky to be using Recall on a securely encrypted Copilot+ PC than to walk around with a Windows 10 laptop that doesn’t use BitLocker or another encryption method.

Can other people on my PC see the snapshots?

The Recall data is stored separately for each user account on a PC. That means even if you share a PC with other people, they won’t be able to look through your Recall snapshots — not unless they can sign into the computer with your user account and credentials.

Will Recall store financial account numbers and passwords?

Microsoft originally said that “Recall does not perform content moderation.” If a password or financial account number is visible on your screen, Recall would save it.

But Microsoft changed its mind. Recall will now filter sensitive information like passwords, credit card details, and social security numbers by default. However, you have the choice here: You can head to Settings > Privacy & security > Recall & snapshots and turn off the “Filter sensitive information” option if you want to see this information in your snapshots.

Either way, Recall won’t capture most passwords you type, since most websites “cloak” password entry dialogs by displaying them as ****. And you can choose to filter out specific websites (like financial websites), use private browsing, or even filter out entire applications to have Recall ignore them. Also, you can delete Recall snapshots at any time.

Any such data won’t appear in your Recall snapshots by default. If you choose not to filter it, it’s critical to remember that only someone who has physical access to your PC — and who can sign into your user account — can access this information. And someone with physical access to your PC can do much worse, including installing malware.

Recall can reopen documents and websites for you — but you can get a lot of information just from the screenshots Recall stores, too.

Microsoft

But couldn’t someone else with access to my PC snoop on it?

To snoop through your snapshots for your private information, people would need both physical access to your PC and to be signed in as you. And, with Microsoft’s changes to Recall, they’d also need to authenticate as you with Windows Hello. Even if you stepped away from your PC and left it signed in, they couldn’t get access to your Recall data without biometric identification or a PIN.

To be fair, even the possibility of that happening does raise concerns. An abusive partner or family member could dig through the snapshots to find private information, for example — if Recall snapshots were enabled and they knew the Windows Hello PIN or were granted access.

However, this was always a risk. That same person could use their access to install a keylogger and remote-monitoring software to snoop on their partner’s PC usage, with or without Recall. Someone you give momentary access to your PC could pull up your email or search for sensitive financial documents. The Recall feature introduces a new way for people to find sensitive information if they already have access to a PC — but, again, they could do a lot of damage even without Recall in the mix.

How concerned should I be about Recall?

It’s clear why Recall is concerning: It marks a change in the way our computers remember and store information. And it seems like an obvious privacy problem if people with access to our PCs can use “AI-style” plain language search to dig through our saved PC history.

In other ways, it’s not a change: It’s a disabled-by-default feature you can choose to use. Even if you do use it, all the data is stored on your PC, so it’s arguably more private than many of the cloud-connected services we use every day.

Critically, Recall doesn’t send any of this data over the internet. There are already many other details we’re giving to Microsoft and other corporations. If you’re worried about the information those companies are receiving about you, Recall isn’t the problem — but there are a lot of other Windows and web features that might be.

Recall could be a big productivity boost for a lot of workers, helping them dig through all the information they’ve seen on their work PCs. If you also use Discord to chat while working, you could filter out Discord and ensure Recall doesn’t capture anything you say in there while it takes snapshots of all the Word documents, Excel spreadsheets, and Outlook emails you go through all day. And, as we’ve been saying, Recall offers a lot of control in general. If you don’t want Recall to capture a browsing session, you can use Private Browsing mode.

Even veteran Windows journalist Paul Thurrott, who is often critical of Microsoft’s privacy practices, has argued that Recall is not a privacy concern. It’s not uploading anything to Microsoft, as he notes — it’s just storing the data on your PC.

But aren’t there still privacy concerns?

While I can see the benefits of Recall — especially for productivity workers who go through a lot of information on their PCs and could save time if they had a faster way to find it — there are some elements of Recall that should give everyone pause. PCs have never captured and stored this kind of information in this way before. It’s a bit of a shock.

Still, Microsoft has made a lot of good changes after the criticism. Disabling Recall by default on business PCs, filtering private information out of snapshots, and requiring Windows Hello authentication to access snapshots are all smart shifts.

But people do have at least some reason to worry about Recall. An attacker with access to a PC could just enable Recall rather than install a keylogger, and then grab private information from the Recall snapshots. That kind of attack could be a little more subtle and difficult to spot than a full install-a-keylogger attack, too. It’s a good thing that this feature will be disabled by default on business PCs.

The most important answers lie ahead

More than anything, we’ll have to see how the risks shake out in the real world. When I first broached this subject, I suggested Microsoft do more filtering of private information and make efforts to protect Recall snapshots from people with access to a PC. Microsoft made those changes.

Perhaps Recall will make everyone realize the risk of giving other people access to their PCs — something that was always a risk when sensitive documents, emails, and browser histories are just a few clicks away.

Of course, Microsoft’s big Copilot+ PC push is about more than AI. The PC industry now finally has thin-and-light laptops with incredibly long battery life to compete with MacBooks. That’s huge.

Even if you disable Recall and turn off every AI-based feature on those new Copilot+ PCs, they’re a big battery life upgrade over your current laptop.

Interested in learning more? Watch this column and sign up for my free Windows Intelligence newsletter to keep up with all the latest intel. You’ll also get three new things to try every Friday and free copies of Paul Thurrott’s Windows Field Guides as a special welcome bonus.

A US federal judge indicated that the trial addressing the Department of Justice’s proposals to curb Google’s dominance in online search will proceed as scheduled in April 2025, even if DOJ officials under President-elect Donald Trump seek to revise the remedies.

The move signals the urgency in resolving the case, which could lead to a historic shake-up of the tech giant’s operations, Reuters reported.

US District Judge Amit Mehta, overseeing the case in Washington, confirmed Tuesday that the trial will proceed as scheduled despite potential changes in DOJ leadership and priorities under President-elect Donald Trump’s administration.

“If there is going to be a re-evaluation of the remedies that are being requested, it needs to be done quickly,” the report said quoting Judge Mehta from a hearing.

The DOJ has proposed sweeping remedies to curb Google’s influence, including a forced divestiture of its Chrome browser and potentially its Android operating system. Both products serve as critical distribution channels for Google Search, a service found to operate as an illegal monopoly in Mehta’s August ruling.

For context, Trump is likely to pursue ongoing lawsuits against Big Tech, several of which originated during his first term. However, his recent remarks expressing skepticism about a potential Google breakup underscore the significant influence he will wield in shaping the direction of these cases.

“If you do that (splitting Google), are you going to destroy the company? What you can do without breaking it up is make sure it’s more fair,” Trump said at an event in Chicago in October.

The Justice Department first filed the antitrust lawsuit against Google in 2020, during former President Donald Trump’s first term. It accused Google of leveraging its dominance in search and advertising markets to stifle competition.

In August 2024, Mehta ruled that Google violated US antitrust laws, setting the stage for the ongoing debate over remedies. The DOJ under President Joe Biden proposed additional measures, including requiring Google to share search data with competitors, limiting investments in rival technologies, and restricting acquisitions of companies in search or query-based AI.

Google’s pushback and trial stakes

Google has sharply criticized the DOJ’s proposals, calling them “staggering” and warning that they could harm American technological leadership. The company argued that measures like forced divestitures and data-sharing mandates could weaken its competitive edge and disrupt the broader digital ecosystem.

“We’ve invested billions of dollars in Chrome and Android. Breaking them would change their business models, raise the cost of devices, and undermine competition with Apple,” Google said in a blog post in October.

The upcoming trial is expected to highlight the role of artificial intelligence in reshaping the online search landscape. Prosecutors plan to call witnesses from major AI players, including OpenAI, Perplexity, Microsoft, and Meta Platforms, to underscore the competitive challenges Google’s practices pose to innovation.

Political dynamics and antitrust implications

President-elect Trump, who expressed skepticism about a Google breakup, has yet to outline his administration’s stance on the DOJ’s proposals. However, Judge Mehta’s decision to maintain the trial timeline suggests limited patience for political recalibrations.

The case represents the most aggressive antitrust action against a tech company since the DOJ’s unsuccessful attempt to break up Microsoft two decades ago. If successful, the remedies could significantly alter the competitive dynamics in online search and advertising markets while setting a precedent for regulating the tech industry.

The trial remains a critical test of the federal government’s ability to rein in Silicon Valley’s most powerful players in an era where technology increasingly shapes global markets.

Nvidia on Monday showed off a new generative AI (genAI) model that can be used to create all kinds of sounds and music: Fugatto (which stands for Foundational Generative Audio Transformer Opus 1).

By entering a text prompt, a user can make Fugatto create basically any sound, such as a trumpet barking like a dog. The genAI ​​model can also be used to change the dialect of a singer or turn a piano piece into a song, according to Reuters.

Fugatto has been trained on open source data, and there is currently no official release date. However, the idea is that the ​​model will eventually be used in the production of music, films and games.

The video below from Nvidia highlights some of Fugatto can do.

Brazil’s antitrust body has joined a chorus of regulators to demand that Apple permit external payment methods in iOS apps. It’s just the latest page in an ongoing story, but might be enough to break this camel’s backbone. 

What this means, at the risk of stating the obvious, is that Apple now faces so much pressure to open up to external payment systems, it could finally make sense for it to bite the bullet and open up across all its territories rather than continue to fight. 

Apple has already been forced to open up in this way — and also to third-party app stores — in the EU, and to let US developers sell in-app content outside the App Store. It now faces similar pressure in the UK. But while it resists each of these moves, there is a cost to the company in legal fees and reputational damage attached to each battle in this conflict — at some point, it might make better business sense to cede the field.

A potential opportunity

While I don’t expect Apple is at all thrilled at how these cookies are crumbling, perhaps there is a way to turn all of this adversity into opportunity. If there’s ever been a time to add features and improvements to the payment systems Apple already provides, this is it.

It might also be time for Apple to take its payments infrastructure to other platforms and markets. Why shouldn’t you be able to pay for Android apps using Apple’s payment systems? Why not offer Apple payments to gamers from within Fortnite? Why not turn payments into products and grab an Apple-sized slice of the wider payments pie?

Customers from inside other ecosystems might be ready to embrace Apple’s rock-solid, highly secure, privacy-first payment system. What I’m saying is that Apple has a unique chance to compete, one from which it can continue to evangelize the advantages of the services (and platforms) it already provides for in-app purchases and everything else.  At the end of the day, the best way to keep people using its payments systems is to convince them that they want to use that system — even if they have a choice of others to use. 

With choice being imposed on the company, the company has an excuse to compete right back at competitors.

Who will lose?

Apple will not be blind to this, but support for external payment systems on its platforms remains very new and is only visible in a small number of markets. Given the potential risks of fraud and worse, it makes sense for Apple (and everyone else) to take a wait-and-see approach to extending this openness to new markets. It is just good practice to monitor what scams, frauds, and other attacks will emerge as third-party services are used on iOS in the EU. It’s not inconceivable that part of Apple’s reluctance to open up more widely yet (other than the money) is a desire to assess the perils and pitfalls of doing so — a trial in which Apple’s European customers are the crash test dummies.

But regulators don’t seem terribly keen to wait and see. Regulators in India, Brazil, UK, US, Japan, South Korea, and elsewhere now seem to agree that Apple must lift restrictions on payment methods for in-app purchases. It’s going to happen in the end.

What price platform integrity?

Even then, another problem Apple faces in that is that each nation could demand slightly different approaches to lifting those restrictions. The problem is that there is a development and infrastructure support cost, not to mention legal expenses, to each of those dictated approaches. What that means is that the less harmonious Apple lets payments on iOS become, the higher the cost of business. 

To avoid weakening the platform with a thousand cuts, it just makes more sense to lift the restrictions internationally, while also putting in place firm safeguards that permit Apple to swiftly remove any payment services identified as fraudulent or lax in security from its platforms. 

Now, I’m on the record arguing that I think there is a very high probability that once payment systems in apps are opened up this way we will see fraud, identity theft, and other forms of financial crime affect against Apple’s so-far highly secure platform. I think that’s inevitable.

Consumers will be damaged, and in the case of those using non-Apple payment services or app stores they cannot expect to get support from Apple. They may have accessed a non-Apple service on an Apple device, but the exchange will be between them and the service, not them and Apple. There will be confusion and broken hearts. This is what will happen.

Managed decline

But Apple can manage the experience and focus on showing the many ways it offers a better and safer system to use. It also means bowing to the inevitable and building something that satisfies regulators enough that they don’t choose to force Apple to build a system that dilutes its own platform. 

So, why has Apple resisted so much? Perhaps because it knows there are other criticisms reaching the anti-trust runway. Perhaps it feels that it makes sense to put up a fight on this particular hill in order to give it time to shore up better defenses on the other hills it currently holds.

All the same, the judgment coming out of Brazil suggests the company is running out of time to prepare for other battles, and now might be time to concede on this particular point. Despite which, if I were in Apple (or a regulator’s) shoes, I’d still try to delay any such move until the first casualties from the European experiment are identified and lessons learned. 

You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

For decades, enterprises have allowed their workers to use whatever free browser they wanted to access the most sensitive files possible. CIOs believed that security software in the environment — such as endpoint security apps or supposedly secure web gateways — would deliver any needed protections.

And until 2020, that view was somewhat valid. But when various pandemic-fueled changes hit the workplace, almost everything changed. But as extreme browser exposure became far more dangerous, the shift was so gradual that almost no one in IT noticed any danger. Those changes included massive numbers of new remote sites; skyrocketing shifts away from on-premises tools and apps to the cloud; and far more SaaS deployments. 

The browser issue here actually arises from two distinct problems: virtually no limits on which browser can be used and no protections at the enterprise level that sit atop those browsers. 

The first is the most bizarre. 

Somehow, IT permits any browser to be used in their sensitive environments. Can you imagine that being permitted for anything else? How many CIOs would tell workers they can use whichever VPN app they want, including free consumer-grade VPNs? Would an enterprise CIO be OK with someone in finance ignoring the corporate license for Excel and instead opting to put sensitive payroll details into a freeware spreadsheet found at a gaming site in China? Or maybe an employee could forego a company-paid Zoom account for discussions of that upcoming acquisition and use a freebie service no one’s ever heard of? 

[Related: 10 tips for a secure browsing experience]

IT typically maintains strict controls over all software that touches their privileged areas, but browsers are a security free-for-all?

Let’s delve briefly into the history. When graphical browsers first moved into the enterprise in large numbers (don’t forget that the earliest browsers, such as Cello and Lynx, were pure text) around 1994, the goal was to make it as easy as possible for people to interact with the web. The internet at that point had been around for decades, but the web had only recently become popularized. 

The problem is that as environments became exponentially more complex and access to ultra-sensitive data soared, IT didn’t stop to reconsider ancient browser policies. 

If IT admins were to choose one specific browser to mandate, controls would become light-years easier. They could even require users to access the latest version from IT, allowing for updates to be strictly maintained. Internal web pages could be designed for that browser, making it far more likely to deliver an identical experience for all users. 

I routinely run into secure areas where critical text (such as the “next” button) is offscreen. That means trying three or four browsers until one works. Imagine that problem disappearing simply by mandating one browser for all. 

That kind of corporate mandate brings up a few issues:

• Desktop vs. mobile. Some enterprises might need to consider standardizing on one browser for desktop and possibly a different browser for mobile. 
• IT political issues. Some of the browsers with major market share are deeply integrated with one vendor’s environments, such as Google Chrome and Microsoft Edge. Depending on how your environments are integrated with different platforms, this could be an issue. 
• Compliance. Some of the browser makers are more aggressive at pushing privacy and other data boundaries, especially when generative AI is involved. Standardizing on one of those might lead to corporate compliance issues, especially if you have a substantial presence in Western Europe, Australia or Canada. 
• Geography. Beyond the compliance issues, there are language and other regional support issues to consider, especially if you have a major presence in Asia. 

That brings us to problem two. Browsers were never designed to be even a little bit secure in the early days — and not much has changed today. That’s why IT needs to insist that something act as a secure layer between your environment and any browser — even your hand-chosen favorite browser. 

Because the needs of every enterprise are different, there’s no one-size-fits-all browser security solution. The browser security layer must play well with your existing systems and your particular compliance needs — colored by geography and verticals — are critical factors.

“The browser is the number one app that everyone is using. The browsers of today are much more powerful than the older versions,” said Dor Zvi, CEO of security firm Red Access. “They allow you to run Javascript, login and tokens and render HTML. The browser today is so powerful that it acts almost like an operating system.”

Zvi argues that there is a reason those browser capabilities are so dangerous. 

“A lot of the attacks today can now happen entirely within the browser. It is happening inside the frame of the browser, which means it is not on the network side and not on the endpoint side. The browser now holds the cookies and tokens for all of your applications,” he said. “Let’s say someone is trying to steal my Okta two-factor authentication. [The attacker] can run it by solely using the browser privileges and no one will ever know about it.”

Another problem with allowing any browser from around the world to access your systems involves browser extensions. In the same way Apple and Google can’t adequately police their apps to detect and remove malicious ones, browser teams can’t verify the legitimacy of extensions. A malicious browser often has unlimited access to everything the browser can do or see. That’s why standardizing on one browser is important; it allows IT to also rein in browser extensions.

It’s a lot to think about — but preferably not right before bed. 

I recently wrote about how President-elect Donald J. Trump’s actions on AI might affect Microsoft. This week, I’m focused on what his antitrust regulation and environmental plans — and the biggest wildcard of all, his personal vendettas — could do to the company. 

What Microsoft can expect from antitrust lawsuits

Trump believes that the less regulation on big business, the better. So you would expect him to put an end to antitrust suits against the tech industry. But that’s not necessarily the case.

There’s no doubt that Lina Khan, the head of the US Federal Trade Commission (FTC) who has aggressively pursued antitrust prosecutions against tech, will be let go after Trump’s election. And many of Trump’s advisers, notably venture capitalist Marc Andreessen, would like to see tech antitrust prosecutions to stop. 

However, some advisers close to Trump, including Vice President-elect JD Vance, want the administration to take on Big Tech — mainly because they want to stop Meta and other social media companies from policing against misinformation, white supremacism, public-health health deceptions and election lies.

Microsoft has largely been spared Khan’s prosecutions, even as the Biden administration has targeted Google, Apple, Meta, Amazon, and Apple. The one recent federal antitrust action against Microsoft by the FTC, for buying the gaming giant Activision, didn’t go well for the feds. A judge let the purchase go through, although the FTC has since appealed the case.

That might make you think that Microsoft is in the clear under Trump. But The Washington Post reports the FTC will be investigating Microsoft’s cloud business for anticompetitive practices. In addition, the FTC appeal of the Activision case still stands, so that case could be revived.

Trump could demand that whomever he appoints to head the FTC drop those actions. Odds are, he won’t, thanks to his main tech adviser, entrepreneur Elon Musk. His AI startup, xAI, competes directly with Microsoft, and is now valued at $50 billion after investments this spring from Andreesen and others. Musk also recently amended an antitrust suit he filed against OpenAI, adding Microsoft as a defendant. 

Don’t be surprised if the FTC under Trump not only follows through on Khan’s investigations of Microsoft, but also files an AI suit against the company, thanks to Musk’s influence.

Trump, Microsoft, and climate change

Trump believes climate change is a hoax. He’s vowed to tear up environmental regulations and attack green energy. His campaign slogan, “Drill, Baby, Drill,” and his close friendship with the oil industry make clear that he’ll do everything he can to increase reliance on fossil fuels and kill clean sources of electricity.  

He was also a booster of nuclear power during his first administration, though he wasn’t quite as enthusiastic about it on the campaign trail. Even so, the stock market price of nuclear-power-related companies jumped the day after his election, and most people expect him to be a nuclear backer.

What does this have to do with Microsoft? Plenty. Microsoft has vowed to make itself carbon-negative by 2030, and Trump’s attack on green energy will make it more difficult for the company to find clean energy sources.

Exacerbating Microsoft’s climate-change challenges is the fact that data centers that power AI require a tremendous amount of electricity. As I’ve noted before, Microsoft might be abandoning its promises to fight climate change because of that. And the company could also pour billions into reviving nuclear energy with a proposed deal to reopen Three Mile Island, the site of the worst nuclear power disaster in US history. 

Given Trump’s views about climate change and his support for AI, he’ll most likely do everything he can to give Microsoft and other AI companies all the electricity they want no matter the effect on the environment. And he’ll also likely let them go full speed ahead with nuclear power. In fact, Microsoft President Brad Smith recently said he expects Trump to cut environmental regulations to provide Microsoft with all the electricity it wants for its AI data centers. 

Gregory Allen, director of the Wadhwani AI Center at the Center for Strategic and International Studies — he worked on AI at issues the Department of Defense during the Trump and Biden presidencies — agrees. On a call hosted by The Information, he said Trump “can invoke emergency powers and waive a lot of environmental regulations to allow people to build new nuclear and other electrical generation capacity in order to power the big data centers that folks want for these advanced AI models.”

He added that he expects that to happen “pretty early in the Trump Administration.”

Trump’s vendettas and grievances

The president-elect is driven by vendettas and grievances more than he is by policy. And when it comes to tech, he has plenty of them.

In the 2020 election, Meta founder Mark Zuckerberg and his wife started a foundation “to ensure that everyone can vote and every vote can be counted.” Since then, Trump threatened to investigate him and send him to jail if re-elected, saying, “We are watching him closely, and if he does anything illegal this time, he will spend the rest of his life in prison.” 

Zuckerberg got the message, offering accolades, saying after last summer’s assassination attempt, “Seeing Donald Trump get up after getting shot in the face and pump his fist in the air with the American flag is one of the most badass things I’ve ever seen in my life…. On some level as an American, it’s like hard to not get kind of emotional about that spirit and that fight, and I think that that’s why a lot of people like the guy.”

Then there’s Amazon founder and Washington Post owner Jeff Bezos. When Trump was president, he frequently took aim at Amazon and Bezos because the Post published articles that angered Trump. He didn’t just criticize and threaten him; Trump also yanked a multi-billion-dollar cloud contract with the Defense Department from Amazon.

This time around, Bezos is doing Trump’s bidding. He canceled the Post’s planned endorsement of Vice President Kamala Harris even though the newspaper has endorsed candidates for president for decades. After Trump was elected, Bezos praised him, writing on X, “Big congratulations to our 45th and now 47th President on an extraordinary political comeback and decisive victory.”

Those are just two of tech titans who have praised Trump even though he had targeted them. Microsoft CEO Satya Nadella has so far managed to avoid getting on Trump’s bad side. He hasn’t gone out of his way to praise the president-elect, either, offering Trump only a pro forma congratulation after the election.

But with Musk as a Trump adviser, and what will likely be a big focus on AI in the new administration, it’s not clear whether Nadella will be able to stay out of Trump’s crosshairs. What’s also not clear is how Nadella will react if Trump threatens him — and how that might affect Microsoft’s financial future and its sense of itself as a moral company.

The “robust computer that’s very, very tiny” — introduced by Apple CEO Steve Jobs almost 20 years ago — just got even tinier. And once again, if you’re thinking of switching from Windows, there’s little excuse not to climb aboard; the “most affordable Mac ever” is also among the fastest consumer AI desktops money can buy.

While the Mac mini in hand is considerably smaller, its cost increased just a little and computational performance improved exponentially. These impressive changes allow the it to be a gateway for switchers, a second computer for any mobile Mac user, and a highly capable desktop for everyone else.

It’s also a server, a computer to which you can offload big tasks and it’s quite capable of handling the kind of cutting-edge productivity software you might use on a MacBook Pro, though perhaps not as efficiently. 

In the interests of objectivity, I should say up from I love the new Mac mini. It’s a triumph, a culmination of everything the first Mac mini aimed to be, but much, much better. Introduced along with the also superb MacBook Pro, Apple’s Mac line-up proves that, with Apple Silicon inside, the company is a the top of its game.

What you can expect under the hood

All this capability comes because of the amazing M-series processor Apple has slotted inside and reflects the device’s extensive processor history that straddles the company’s PowerPC chips on its first release, the Intel years, and today’s super-efficient, low-power chips that put Apple ahead of the industry. There’s a lot to love, starting at $599 (though the M4 Pro with 14‑core CPU and 20‑core GPU, 48GB, and 1TB SSD model I tested costs a lot more, $2,199.) That price tag might dent the superlatives a little, but probably not fatally. 

For a company made famous by the quality of its design, the Mac mini you see today isn’t a major departure from the models of yesteryear, other than size. This third major redesign remains faithful to the breed — a compact all-in-one metal box designed to work with the mouse, keyboard and display you already own. Now just 2-in. high, the 5-in.-by-5-in. (100% carbon neutral aluminum) box remains, resolutely, a Mac mini.

Such is the classic simplicity of Apple design, if you’d been abducted by aliens two decades ago and taken to the peaceful planet Zog to hang out with and learn from an enlightened species, you’d still recognize this as a Mac mini when you returned. (Though you’d probably be disappointed at the state of enlightenment here on Terra Ferma.)

But alien adventures aside, because it aims to work with kit you already own, connectivity has always been important to the mini. The new model offers two USB-C ports, HDMI, Gigabit Ethernet, three Thunderbolt 5 ports, a headphone jack, Wi-Fi 6E, and Bluetooth 5.3 — though you no longer get USB-A, putting that standard even further back in history. You also don’t get an SD card slot, but you didn’t in the last model, either.  You can now drive up to three external displays, which is amazing, really, and I bet many of us take that for granted.

The power button (which you rarely, if ever, need to touch) is on the lower left corner of the 1.6-pound device; that positioning raised many critical cat calls when it was spotted, but if that’s all the critics have then Apple has got something right.

What it does

Apple says the Mac mini with M4 Pro is up to 20x faster than the fastest Intel-based Mac mini. The benchmark results I got back that assertion up, and more. I was a little open-jawed at the results I got and had to run tests multiple times they impressed me so very much.

Time for some benchmarks:

Geekbench 6.3

• Single Core: 3,8715.
• Multi Core: 22,314.
• OpenCL: 69,013

The CPU results are incredibly impressive. If you check the Geekbench Mac charts, you will find they mean the Mac mini delivers at least as much punch as the currently available Mac Studio, or last year’s 16-in. M3 Max MacBook Pro. There is no performance compromise whatsoever in this machine.

Cinebench R23

• 22,737 CPU multi core (a top three position, up there with Intel Xeon W and AMD Ryzen Threadripper 2992WX).
• 2,137 CPU single core (leader of the pack).

Valley

FPS 101.3 

It is important to note that Valley isn’t optimized for Apple Silicon and relies on Apple’s Rosetta technology, so it’s not a fair comparative test. But it does illustrate just how performant these little Macs have become.

You’ll find additional benchmark tests at MacStadium, where new M4 Mac minis are already being put into service as servers in real-life, mission-critical environments. They note that the M4 Pro, “tears past all the previously available Mac mini models, and even puts some of the older Studio models to shame.” 

You’ll find a similarly fabulous statement from an impressed Jeff Geerling, who says: “The chip isn’t the fastest at everything, but it’s certainly the most efficient CPU I’ve ever tested. And that scales down to idle power, too — it hovers between 3-4W at idle — which is about the same as a Raspberry Pi.” 

It is worth noting that most of the time the power efficiency means it will barely feel warm to the touch, no matter how hard you push it. These results, and those of all the other M4-powered Macs, absolutely illustrate the extent to which the shift to Apple Silicon has turned the processor industry upside down, putting once last-place Apple in bidding distance for the throne.

Take it anywhere

The Mac mini is small. You can put it anywhere you need it — on a bookshelf, certainly under a reception desk, anywhere in an office, and in almost any situation where you might need a computer on warehouse or factory floors. The front-mounted USB-C ports and headphone jack make its usage flexible, too. While it is not and nor is it intended to be a portable device, it is worth noting that so long as you have a keyboard, mouse, and display wherever you intend to go, the Mac mini is a computer you can take with you.

What about Thunderbolt 5?

Apple celebrated the introduction of Thunderbolt 5 on these Macs when they were announced. All the same, for most users it means very little. Sure, if you use a compatible Thunderbolt 5 cable and a compatible device, you’ll get data transfer speeds of 120Gbps, but right now those who have those things skew toward being pro gamers and video professionals. That will change of course as Thunderbolt 5 proliferates and becomes cheaper, though it is nice to know that you can use this tiny Mac to power multiple 6K displays.

Thunderbolt 5 will also be important to those who choose to use the new macOS feature that lets them use larger Mac apps that are stored on external SSD drives.

Time to upgrade?

The new model stacks up proudly against Apple’s first M1-series Mac mini. You’ll see significant performance gains, and while the M1 Mac mini I’ve used as my daily drive ever since it was introduced has never let me down, I did experience a perceptible difference in performance.

Four years later, is it time to upgrade? I think it might be, and the fact I’ve had four trouble-free years with an M1 gives me a lot of confidence to expect more great years with an M4 model.

However, in contrast to the Intel Macs, the question of whether or whether not to upgrade shouldn’t be a question at all — of course, you should. The difference in performance was like night and day when the M1 models first appeared; with the M4 series, you’ll feel like you just swallowed a glass of iced water in hell, as someone once said.

Unlike the performance compromise Mac mini represented back in the day, with Apple Silicon you can look forward to pro performance at a price that’s more within reach. 

A dream realized

The thing about the price is important. It’s hard to ignore a computer that starts at $599 and can kick out this level of performance. As a desktop, it ticks most boxes:

• Windows switchers will like that they might be able to continue using existing kit with the system, and they’ll like it even more once they realize these Macs are so powerful they’ll run Windows better in VM mode than some PCs. 
• Pro users will quickly find these Macs are capable of pro level performance that matches or exceeds some of last year’s more expensive Mac models.
• Enterprises can be confident that these machines can be deployed across a wide array of situations and handle their tasks really well.
• And every Mac mini user will appreciate that there is enough processor “oomph” inside these devices that we will still be enjoying a great experience using them in three, four, five or more years’ time. As mentioned above, my M1 Mac mini has never missed its stride and is four years old.

With its new – and still unmistakably Apple Mac mini design — the new model looks good, is whisper quiet, runs almost every application you might want to run, and demands hardly any desk space. If you need an Apple desktop or need to put an Apple system together at as low a price as possible, then the great thing about these Macs is you won’t feel at all compromised – these things shift!

All in all, this is a triumph, an absolute accomplishment of the journey Apple set out on when the first ever Mac mini models appeared. I can’t recommend it enough. 

You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

Microsoft has released a new collection of tools and a guide to fix security issues that have arisen around the way the company’s generative AI (genAI) Copilot assistant handles information. Namely, the tool’s indexing of an organization’s internal data can lead to the AI ​​assistant sharing sensitive information when it shouldn’t.

A Microsoft employee familiar with customer complaints tells Business Insider: “Now, when Joe Blow logs into an account and starts Copilot, they can see everything. All of a sudden, Joe Blow can see the CEO’s email.”

Business Insider reports that the behavior prompted several organizations to delay using Copilot for security reasons. “Many data governance challenges associated with AI were not caused by AI’s arrival,” a Microsoft spokesperson told the publication.

Instead, according to the spokesperson, AI tools like Copilot highlight how companies need to take proactive responsibility for how they manage internal documents and other information.

After a series of delays, Microsoft’s Windows Recall feature is now available in a limited preview for Copilot+ PCs. 

Recall, which takes regular snapshots of a user’s screen to provide a searchable timeline of actions, drew criticism from security and privacy experts when it was unveiled earlier this year. 

On Friday, Microsoft announced a new Windows 11 build for Windows Insider’s Dev Channel that includes Recall and Click To Do, another Copilot+ feature in preview that provides an interactive overlay on a user’s screen. 

The number of Windows Insiders testing the features is likely to be small at this stage, as Windows 11 Insider Preview Build 26120.2415 (KB5046723) can only be accessed on Copilot+ PCs with a Qualcomm Snapdragon chip. Support for devices running AMD and Intel chips is “coming soon,” Microsoft said in a blog post. 

 The announcement marks the next step toward a full release for a feature that was labeled a “privacy nightmare” upon its announcement in May. 

A planned rollout in June was postponed in response, and Microsoft has attempted to address security and privacy concerns with several updates. This includes making the feature “opt-in,” requiring biometric authentication with Windows Hello prior to use, blocking detection of personal details such as credit card details and passwords, and the addition of a “virtualization-based security enclave” (VBS Enclave) to secure data on a user’s device.

The US government has scaled back Intel’s preliminary CHIPS Act grant from $8.5 billion to under $8 billion, reflecting concerns over the company’s delayed investments and financial woes, The New York Times reported. The funding was part of the government’s effort to boost domestic semiconductor manufacturing amid growing global competition.

Intel, originally seen as the largest beneficiary of the CHIPS Act, has struggled to meet expectations following its biggest quarterly loss in its 56-year history. The cut coincides with a $3 billion military contract offered to Intel to produce chips for the Department of Defense, the report said citing sources who did not wish to be identified.

In March 2024, the Biden administration and Intel signed a preliminary memorandum of terms (PMT) for an $8.5 billion funding package. This support was part of Intel’s broader plan to invest over $100 billion in expanding its US manufacturing operations, including the construction of new chip facilities in Arizona, Ohio, Oregon, and New Mexico.

The agreement also included up to $11 billion in additional loans from the US government, aimed at strengthening Intel’s position as a key player in the evolving AI-driven semiconductor landscape.

The decision to reduce the grant underscores the challenges Intel faces as it attempts to reclaim technological leadership while fulfilling the US administration’s vision of revitalizing domestic chip manufacturing.

However, there is no clarity on the other terms and conditions of the reduced grant package.

Investment delays and strategic setbacks

The funding reduction comes as Intel pushes back the timeline for completing its Ohio chip manufacturing project from 2025 to the end of the decade. The delay, coupled with persistent challenges in matching the technological advancements of rivals like Taiwan Semiconductor Manufacturing Company (TSMC), has dampened confidence in the company’s ability to deliver on its commitments.

“The delay in Intel’s investment is especially concerning given the current surge in demand for chips, driven by the rise of AI,”  said Rachita Rao, senior analyst at Everest Group. “With AI transforming the industry, the existing IT infrastructure is becoming insufficient to handle its processing requirements.”

Intel’s difficulties come as the Biden administration seeks to reduce US reliance on Asian supply chains through the CHIPS Act, a $39 billion initiative aimed at boosting domestic chip production. In March, President Joe Biden highlighted Intel’s role in transforming the semiconductor industry during a high-profile visit to Arizona.

However, Intel’s setbacks now present significant hurdles to achieving that vision, the report noted.

Oversight and milestones

Commerce Department officials, tasked with ensuring accountability for CHIPS Act funding, have set stringent performance benchmarks, such as building plants, producing chips, and securing customer commitments for domestically made products.

Intel’s struggles to meet these goals reportedly complicated its negotiations for the final grant terms, according to the report.

Meanwhile, TSMC has secured a $6.6 billion grant under the program while committing over $65 billion of its own funds to US factory construction.

“Additionally, Intel is pursuing riskier strategies at a time when TSMC is focusing on a low-risk, high-production model that appears to be yielding strong results,” Rao noted. “Given Intel’s inability to effectively compete in the current market, the reduction in funding seems justified to some extent.”

This, certainly, is not a piece of good news for Intel which has been grappling with significant financial challenges at the moment. The company reported an 85% year-on-year decline in profits and announced plans to cut 15,000 jobs recently. Additionally, the financial downturn has prompted Intel to suspend dividend payments.

The path ahead for US chip manufacturing

The Biden administration viewed the funding as a strategic initiative to lessen reliance on foreign semiconductor supply chains. The US has highlighted the program’s success in driving factory construction, pointing out that the country will soon host facilities from all five of the world’s leading chipmakers.

“Intel is struggling to keep pace with its competitors, particularly TSMC, which dominates the market with its competitive pricing and significant market share,” Rao said.

Intel’s success is vital not just for the company, but for the broader US semiconductor ecosystem. As AI is poised to drive future demand for advanced chips, Intel’s manufacturing capabilities and technological innovations will be crucial in ensuring the US remains competitive in the global market.

However, the reduction in Intel’s grant underscores the challenges of balancing federal investments with corporate accountability. A query to Intel remains unanswered.

Virtually every PC manufacturer has announced, or is already producing, machines with embedded artificial intelligence (AI) functionality. The question is: Why?

Generative AI (genAI) for consumer use already exists through any number of cloud-based services, from OpenAI’s ChatGPT to Google’s Gemini and others.

Even so, next year will be “the year of the AI PC,” according to Forrester Research.

The research firm defines an AI PC as one that has an embedded AI processor and algorithms specifically designed to improve the experience of AI workloads across the central processing unit (CPU), graphics processing unit (GPU), and neural processing unit, or NPU. (NPUs allow the PCs to run AI algorithms at lightning-fast speeds by offloading specific functions.)

“While employees have run AI on client operating systems (OS) for years — think background blur or noise cancellation — most AI processing still happens within cloud services such as Microsoft Teams,” Forrester explained in a report. “AI PCs are now disrupting the cloud-only AI model to bring that processing to local devices running any OS.”

AMD, Dell, HP, Intel, AMD, Apple, Nvidia, and Lenovo have all been touting AI PC innovations to come over the next year or so. Those announcements come during a crucial timeframe for Windows users: Windows 10 will hit its support end of life next October, giving them a real reason to upgrade to Windows 11 — and buy new hardware.

Gartner’s latest worldwide AI PC shipment forecast projects a total 114 million units in 2025, an increase of 165.5% from this year. Key findings in the forecast include:  

• AI PCs will represent 43% of all PC shipments by 2025, up from just 17% in 2024. 
• The demand for AI laptops is projected to be higher than that of AI desktops, with shipments of AI laptops accounting for 51% of all laptops in 2025.
• By 2026, AI laptops will be the only choice of laptop available to large businesses, up from less than 5% in 2023.

“The debate has moved from speculating which PCs might include AI functionality, to the expectation that most PCs will eventually integrate AI NPU capabilities,” said Ranjit Atwal, senior director analyst at Gartner. “As a result, NPU will become a standard feature for PC vendors.”

As the PC market moves to AI PCs, x86 processor dominance will lessen over time, especially in the consumer AI laptop market, as Arm-based AI devices grab more share from Windows x86 AI and non-AI laptops, according to Atwal. “However, in 2025, Windows x86-based AI laptops will lead the business segment,” Atwal said.

But why bother embedding AI algorithms in a computer’s firmware or software — thus, requiring more expensive processors to power them — when you can access those same tools on the web? According to Tom Butler, Lenovo’s executive director of worldwide commercial product management, AI will fundamentally transform PCs, making them not only smarter but also more responsive and secure.

“We see AI-enabled PCs evolving to provide more personalized, adaptive experiences that are tailored to each user’s needs,” Butler said. “The rise of generative AI was a pivotal moment, yet reliance on cloud processing raises concerns around data privacy.”

Each component of a PC plays a unique role in making AI tasks efficient, but the NPU is key for accelerating AI computations with minimal power consumption, according to Butler. In general, he said, AI PCs assist in or handle routine tasks to be more efficient and intuitive for users without the need to access an external website or service.

Apple, for example, last month announced an updated iMac powered by its new M4 chip with an NPU core and Apple Intelligence, an AI-powered assistant that can help users write emails or other content. (More intensive or complex tasks can still be handed off to OpenAI’s ChatGPT. Apple also unveiled M4-powered MacBook Pro laptops and Mac minis — all while touting their strength in handling AI-related tasks.

AI PCs can also boost productivity by handling routine tasks such as scheduling and organizing emails, and by enhancing collaboration with real-time translation and transcription features, according to Butler.

Intel Corp.

Depending on the device, AI technology can also seamlessly integrate with cloud and edge computing for real-time data processing, enabling faster and more informed decision-making. AI-enabled PCs also increase device security by automating threat detection and adapting to new threats as they arise.

For example, Butler said, Lenovo’s Smart Connect enhances device synergy, allowing users to transition seamlessly between Lenovo devices, while ThinkShield provides security across the ecosystem, protecting users in real time.

AI-powered PCs however, generally require more RAM to handle advanced tasks. Apple, for example, is moving from a minimum of 8GB of RAM to 16GB.

Lenovo took a slightly different approach in its RAM support of AI tasks. The company’s “Smarter AI for All” tries to match the complexity of tasks to processing needs. For example, 16GB is suitable for lighter AI tasks when combined with a more powerful NPU, while 32GB or more is suited for users handling complex applications, large language models, or deep learning.

“Users working within AI development spheres will most likely require more RAM, combined with powerful GPU and CPU to ensure low latency and AI model fine-tuning capabilities,” Butler said.

Guayente Sanmartin, HP’s president of commercial systems and displays solutions, said PCs with NPUs capable of more than 40 tera operations per second (TOPS) are transforming daily workflows, with new use cases emerging almost every day.

Tools like Microsoft Copilot and LLMs like ChatGPT not only quickly summarize documents, draft emails, and schedule meetings, they can generate action items from meeting recaps — boosting productivity and freeing up time during the workday, she said.

“We are also seeing creative, editing, design, and content creation use cases,” Sanmartin said. “AI now enables you to generate images, videos, and music just from text prompts — making anyone a composer or designer. A product marketer can now create copy and fully realized assets for a campaign in hours instead of weeks without waiting on creative agencies. Professional video and design professionals also benefit from tools that automate complex processes like background removal or generative expansion with just a single click.”

Also emerging is the ability for AI-enhanced devices to perform language translation in real-time meetings, with live captioning and voice-to-text and voice command, which can help break down barriers between global teams, he said. And AI tools like Virtual Sapiens can provide real-time coaching during meetings to help users come across more professional and communicate more clearly.

A key advantage of running AI on the device instead of using a cloud service is the automatic increase in security. “AI also automates your control over the PC itself,” Sanmartin said. “You can ask AI to check security settings, configure cameras and microphones for calls, as well as adjust your user interface based on your preferences or adapt automatically based on usage patterns. To improve your security and threat protection, AI can monitor your data and behaviors for unusual patterns and stop threats in real time.”

An AI PC that supports local inferencing — running language and visual models directly on the device — has several advantages, according to Sanmartin. It allows users to handle sensitive information securely, as personal, proprietary, or financial data stays local rather than being uploaded to a cloud-based AI, Sanmartin said.

“AI on your PC can leverage the data and content in all your personal files for generative tasks without exposing it to external servers,” she said.

Could AI make things harder?

Ironically, though, the results of a new survey and study conducted by Intel found that current AI PC owners spend more time on tasks than people who use PCs without AI technology. The survey of 6,000 consumers from Germany, the UK and France, found about 53% believe AI-enabled PCs are only good for “creatives or technical professionals.” And 44% see AI PCs mainly as “a gimmick or futuristic technology.”

In all, the survey showed that, in general, users spend 899 minutes cumulatively, nearly 15 hours, on computer-related chores weekly. Intel’s study showed that current AI PC owners spend longer on tasks than their counterparts using traditional PCs because many spend “a long time identifying how best to communicate with AI tools to get the desired answers or response.

“Organizations providing AI-assisted products must offer greater education in order to truly showcase the potential of ‘everyday AI,'” Intel argued.

Intel Corp.

When its uses are understood, leveraging AI tools to handle repetitive tasks, streamline workflows, or even assist in research can greatly boost productivity, according to a 2023 study by AI safety and research company Anthrophic.

While only 32% of respondents who aren’t familiar with AI PCs would consider purchasing one for their next upgrade, that percentage jumps significantly to 64% among respondents who have used one before.  The survey and study also stated that “early data” suggests AI-enabled PCs can also save users about 240 minutes a week on routine tasks.

The problem is that many AI-PC owners simply aren’t aware of the benefits of AI or don’t know how to access the tools, Anthrophic argued. “Despite AI PCs becoming more available to people, 86% of respondents have either never heard of or used an AI PC. Meanwhile those respondents who already own an AI PC are actually spending longer on digital chores than those using a traditional PC. 

The study concluded that “greater consumer education is needed to bridge the gap between the promise and reality of AI PCs.”

For business-to-business (B2B) purposes, AI PCs offer a promising solution, according to Mike Crosby, executive director and industry advisory service Circana.

Just three of 20 US business sectors defined by federal government, including professional and scientific, finance, and health care, represent nearly 50% of the total AI PC unit sales, Crosby said. “Companies are evaluating these new technologies carefully, weighing the benefits of innovation against the risks to their established environments.”

The upcoming October 2025 sunset of Windows 10 support further amplifies the urgency for AI PCs, with nearly 60-70% of the installed base still on older versions. Microsoft’s extended security update (ESU) offers a temporary reprieve, but Circana expects modernization to ramp up quickly as the deadline approaches.

The announcement that Amazon Web Services (AWS) will be Anthropic’s primary training partner confirms rumors of an even tighter partnership between the two companies.

They announced Friday that Anthropic will use AWS Trainium processors to train and deploy its Claude family of models. Further, as predicted earlier this month, Amazon will invest an additional $4 billion in the startup, making its total investment $8 billion.

AWS is already Anthropic’s primary cloud provider, and the OpenAI rival will now also primarily use Trainium and Inferentia chips to train and deploy its foundation models. Anthropic will also contribute to Trainium development in what the companies call a “hardware-software development approach.”

While it’s unclear whether the agreement requires Anthropic to exclusively use AWS chips, it is a move by Amazon to challenge the likes of Nvidia and other dominant players as the AI chip race accelerates.

“This is a first step in broadening the accessibility of generative AI and AI models,” Alvin Nguyen, Forrester senior analyst, told Computerworld.

Accelerating Claude development

Anthropic, which launched in 2021, has made significant progress with its Claude large language models (LLMs) this year as it takes on OpenAI. Its Claude 3 family comprises three LLMs: Sonnet, Haiku (its fastest and most compact), and Opus (for more complex tasks), which are all available on Amazon Bedrock. The models have vision capabilities and a 200,000 token context window, meaning they support large volumes of data, equal to roughly 150,000 words, or 500 pages of material.

Notably, last month Anthropic introduced “Computer Use” to Claude 3.5 Sonnet. This capability allows the model to use computers as people do; it can quickly move cursors, toggle between tabs, navigate websites, click buttons, type, and compile research documents in addition to its generative capabilities. All told, the company claims that Sonnet outperforms all other available models on agentic coding tasks.

Claude has experienced rapid adoption since its addition to Amazon Bedrock, AWS’ fully-managed service for building generative AI models, in April 2023, and now supports “tens of thousands” of companies across numerous industries, according to AWS. The foundation models are used to build a number of functions, including chatbots, coding assistants, and complex business processes.

“This has been a year of breakout growth for Claude, and our collaboration with Amazon has been instrumental in bringing Claude’s capabilities to millions of end users on Amazon Bedrock,” Dario Amodei, co-founder and CEO of Anthropic, said in an announcement.

The expanded partnership between the two companies is a strategic one for both sides, signaling that Anthropic’s models are performant and versatile, and that AWS’ infrastructure can handle intense generative AI workloads in a way that rivals Nvidia and other chip players.

From an Anthropic point of view, the benefit is “guaranteed infrastructure, the ability to keep expanding models’ capabilities, and showcase them,” said Nguyen, noting that it also expands their footprint and access.

“It’s showing that they can work well with multiple others,” he said. “That increases comfort levels in their ability to get training done, to produce models, to get them utilized.”

AWS, meanwhile, has a “’premiere client, one of the faces of AI’ in Anthropic,” said Nguyen.

From silicon through the full stack

As part of the expanded partnership, Anthropic will also help to develop and optimize future versions of AWS’s purpose-built Trainium chip. The machine learning (ML) chip supports deep learning training for 100 billion-plus parameter models.

Anthropic said it is working closely with AWS’ Annapurna Labs to write low-level kernels that allow it to interact with Trainium silicon. It is also contributing to the AWS Neuron software stack to help strengthen Trainium, and is collaborating with the chip design team around hardware computational efficiency.

“This close hardware-software development approach, combined with the strong price-performance and massive scalability of Trainium platforms, enables us to optimize every aspect of model training from the silicon up through the full stack,” Anthropic wrote in a blog post published Friday.

This approach provides an advantage over more general purpose hardware (such as Nvidia’s GPUs) that do more than what is “absolutely necessary,” Nguyen pointed out. The companies’ long partnership also means they may have mitigated performance optimization advantages that Nvidia has with their CUDA platform.

“This type of deep collaboration between the software and hardware engineers/developers allows for optimizations in both the hardware and software that is not always possible to find when working independently,” said Nguyen.