Microsoft has issued an apology for the global services outage that began on November 12, which disrupted access to Microsoft 365 products for governments, hospitals, schools, and businesses around the world. The corporation says that neither AI nor malicious actors were at fault, but that the outage occurred as a result of an underlying infrastructural issue introduced during a network upgrade.

"We are deeply sorry for the disruption this incident has caused our customers and partners. We know the trust they place in our services and take that responsibility extremely seriously," Microsoft said. ". We are seeing broad recovery and stability across our services."

While popular speculation has been eager to attribute the outage to vibe-coding—and I can't blame anyone for speculating, considering Microsoft's push for company-wide AI adoption—the corporation says the outage was "a technical issue not related to cyber or AI/agents." It's unclear whether another word was meant to follow "cyber." I'm just telling you what the company wrote.

"The incident was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change," Microsoft said, casually brutalizing the syntax of causality. "We have addressed the underlying issue and continue to restore service to remaining customers and regions."

Microsoft acknowledged that the restoration of service was slower than many might have hoped, but said that it brought services back online gradually to ensure stability.

Even if AI wasn't to blame for this week's outage, it's tempting to wonder whether any of the 10,000 employees Microsoft laid off in 2023 as part of its ongoing cost-cutting strategy might have known how to avoid the costly hiccup. Or any of the 6,000 employees it laid off in May 2025. Or the 9,000 it laid off that July. Or the 4,800 it laid off a year later, including 1,600 at Xbox with another 1,600 to follow. Or the 60,000 employees it extended voluntary buyout offers to last month.

Ah, well. No way to know. Microsoft said it will provide a more detailed incident report once its internal review of the outage is complete.

A couple of days ago I reported that Reddit is rolling out reviews for different products, informed by Redditors' thoughts and molded by AI into into dedicated review pages for these products. Given Reddit says "many reviews also include images, prices, and purchase links", I wondered whether Reddit would be making any affiliate money from this. Well, I now have an answer: the company tells us it will not receive a cut.

In an email sent to me, a Reddit rep explains: "Reddit Reviews is an organic experience, and Reddit does not receive a cut of any purchases made through the buy it now button on product pages."

Not that I would have had a massive problem with this—PC Gamer earns affiliate revenue itself—but I suppose it's good to know there won't be a financial incentive there at least in that respect. And it does at least tie in with the aimed "organic" nature of Reddit in general.

I was also concerned that users might review-bomb products they don't like, or that companies could use bots to positively influence reviews for their own products. Reddit reassures us about that, too:

"Platform-wide, we are taking numerous steps to prevent and enforce against inauthentic content."

Some of these steps include:

• (Platform-wide) A layered approach involving moderators, the safety team, and user downvoting.
• (Platform-wide) Improvements to automated tools.
• (Reddit Reviews) LLMs trained to identify comments that describe personal experiences with a product.
• (Reddit Reviews) Safeguards against inauthentic content including limiting repeat contributions from the same account, excluding comments that speculate or offer second-hand claims, and excluding brand-owned account contributions.

That's certainly more reassuring than nothing, but we'll have to see how it goes in practice. AI and bot farms make the modern landscape tricky business, and I'm sure there must already be actors out there giving pretty convincing supposedly first-personal reviews of products on different platforms, whether to bolster or undermine these products.

As always, the proof will be in the pudding (or rather, for the pedants, in the eating of it).

Reddit Reviews is rolling out to US iOS users and will cover around 10,000 products, but the company plans to add more products and expand to other locations throughout 2027.

AI-enabled hacking is only continuing to get more impressive, and more scary, as a French bank claims a hacking group has cost it €100,000,000 (around $110,000,000)—about one quarter's profits. So not much, then?

That's according to an interview with Julien Carmona of Crédit Mutuel Arkéa with the AFP, which Boursorama reported. Carmona detailed the hack in an interview, claiming it to be a "heist of a new kind", stating the company has the "dubious honor of being the first French bank, and I believe the first European bank, to be the target of a successful fraudulent attack." (machine translated).

Carmona says the hack is "unprecedented in terms of sophistication" and "technicality".

Effectively, the hacking group managed to create fake IBANs, including those from Fortuneo (a bank owned by the group), "to trigger payment orders from Crédit Mutuel Arkea's own account to external parties using instant payment systems". These, "by definition have fewer controls than on classic transfers."

Carmona did clarify to Les Echos that though they use Fortuneo IBANs, "Fortuneo's system, and therefore its customers, were not affected."

Reportedly, the hacking group used a compromised service provider of some kind to get privileged access, thereby bypassing restrictions normally in place to stop this kind of hack. When asked by Les Echos why they didn't reveal the name, Carmona said, "This is a collective decision by the banks. This information is not in the public interest and could harm the security of the banking sector. Similar service providers could then be targeted."

Starting on Sunday, November 1, the hacking group apparently made sure "everything was done to circumvent controls, to make these illegitimate transfers appear legitimate."

They then transferred the funds to legitimate European banks, where they then took that money and spent it on crypto platforms. This hack wasn't spotted until Tuesday, November 3, when access to the bank's information system was cut off, but by that point the damage was done.

The French bank has since brought in crypto specialists, who reportedly spotted the abnormal transaction on Sunday and Monday "But, at that time, they themselves, and no one to my knowledge, made the link with Crédit Mutuel Arkea".

Carmona says the hacking organisation was "a very well-organised, very well-equipped criminal group" and "very likely aided by advanced artificial intelligence."

It's not yet clear what role AI played here, but it's particularly good at analysing data, and of course has found value in coding over the last year. AI was only recently used by a cyberhacking gang to target 60 firms a few months ago. In this Korean case, though, the gang only managed to make $210 in ransom payments before being caught.

Welcome to the new era, folks, where we increasingly seem to be heading closer to a daily coin flip over whether our core online services stay live or go down. This time, it's Microsoft. Down Detector reports problems with Microsoft 365 services since about 12:30 pm (ET) yesterday, November 12, and issues are still ongoing.

Countries across the world are affected, and some institutions have been affected severely. Parts of the UK government, for instance, are having problems sending and receiving emails. The UK's health service, the NHS, has reportedly said it is unable to access online consultation for patients. The US and other governments have reportedly been impacted, too.

Microsoft explains that certain features of services "that interact with Exchange Online and email services are impacted." So, it looks like Exchange Online could be the culprit. This is the back-end of Microsoft services that use calendars, tasks, emails, and so on.

Affected services include MS365 admin center, apps, Education features, Copilot, Defender for Office 365, Defender XDR, Fabric, Graph, Intune, Planner, Power Apps, Purview, some elements of Teams, To Do, SharePoint Online, Universal Print, and Windows 365.

In other words, a very extensive list. It's also a list that a bunch of institutions and companies rely on to manage their day-to-day, thus why this is such a big deal.

The latest update from Microsoft, at the time of writing, explains:

"Recovery actions are ongoing and have progressed in phases across affected Microsoft 365 infrastructure. Global telemetry and customer reports show incremental improvement in several workloads; however, service health remains variable and capacity has not yet fully returned to baseline.

"Network and traffic-management actions remain active. To protect email during recovery, we’re continuing to prioritize queue and backlog processing, and the email timeout window has been extended to 48 hours to preserve delayed messages.

"At this time, we have no evidence of malicious activity associated with this incident. We will continue to treat this with urgency until full mitigation, and we will share a full Post-Incident Report with root cause and defense-in-depth actions after mitigation is complete."

That there is "no evidence of malicious activity" as the cause is promising, but it doesn't stop this from being a severe event that's causing a ton of problems in a ton of countries.

This is the big risk of relying on a single kind of service for all your important online dealings. We saw the same risk bear fruit last year with a global AWS outage that took out a load of online services for a while, and recently we saw similarly widespread issues as a result of PHP vulnerabilities.

We're seeing more and more as time goes on that services from big, usually-reliable companies aren't immune to catastrophe. Let's hope Microsoft figures this one out ASAP.

In a message marking the 25th anniversary of the first Xbox console, Xbox CEO Asha Sharma predicts that the future of gaming will look much like its beginning, "when a small team with a big idea could change how the world plays." And even though the term never actually appears in her post, yeah, I'm pretty sure she's talking about AI.

Sharma's message kicks off with a blast from the early days of Microsoft, beginning with a look at Microsoft Adventure, an adaptation of Colossal Cave "that looked more like MS-DOS than an early version of Minecraft," and the first-ever Microsoft staff photo that showcases some absolutely spectacular hair. From there it's a quick trip through Microsoft Flight Simulator, Quake, DirectX, and then in 2001, the original Xbox.

Naturally, there's a tribute to the "generations of employees, studios, developers, and partners" who helped create and grow Xbox, and the people who have played on various generations of consoles over the years.

"Since joining Xbox, I've heard stories of friendships that began over a session and lasted a lifetime, of people who met the person they would marry, and of parents now sharing the games they grew up on with their children," Sharma writes. "I've also heard about people who are no longer with us, and how much those hours spent playing together still mean to those who love them. Ultimately, this is why we exist."

From there, Sharma offers thoughts on the future of Xbox, and this is where I think it gets interesting: After comparing the size and capabilities of the original Microsoft Flight Simulator with the 2024 edition, she continues, "While modern games demonstrate the heights of human craft in ways unimaginable just decades ago, the cost, resources, and risk required to create them have grown just as dramatically. As a result, too many ideas never become games, and too many games fall short of what their creators imagined. For decades, this has been the unavoidable tradeoff of ambition. We must change this."

How exactly that change will come about goes unspecified, but my guess? She's talking about AI.

"Throughout human history, the scale of what we could create was determined by the people, capital, and resources we could command. That relationship is beginning to break," Sharma writes.

"A single person will soon be able to wield capabilities that once required entire nations. A small team will be able to build worlds that today demand hundreds of developers and hundreds of millions of dollars. And our largest studios will be able to pursue ambitions as far beyond today’s games as Microsoft Flight Simulator 2024 is from its first entry in 1982."

The explosive rise in the cost of game development is undeniable. The pioneering open-world sci-fi RPG Starflight was literally made by five guys; its distant descendant Mass Effect: Andromeda credits 14 people for working on "audio design" alone. Take-Two CEO Strauss Zelnick summed up the situation quite aptly earlier this year, calling big-budget game development "a high-stakes game for big boys only"; the development budget of the upcoming Grand Theft Auto 6 is being talked about in terms of billions.

Despite Sharma's optimistic presentation, though, I'm not entirely confident that an AI-powered future where one person—or even a few—can do what currently takes hundreds is really a desirable outcome.

Alongside questions of ethics and artistic vision, there are also very pragmatic concerns about the long-term future of the games industry itself. Because really, the embrace of AI—not just at Xbox, but everywhere—is about cutting jobs. Bethesda had nine concept artists working on Starfield: What happens when somebody in an office decides they can get the same job done, pretty much, with two guys hammering out prompts? Maybe even just one? Quarterly profits and shareholder return may go up, but I doubt we'll be saying the same about the quality of the games we play.

Sharma concluded by writing that she believes "the future of gaming will look more like its beginning, when a small team with a big idea could change how the world plays." I would argue that small teams are already doing that—the big, billion-dollar blockbusters that have prompted these worries seem far more committed to merely giving us more of what we've already got.

Users of Humble Bundle—the pay-what-you-want game-bundle storefront that will also sell you books and software—have been alarmed by a new redesign that seems to have axed a foundational piece of login security tech.

The just-released new Humble Bundle website has no support for two-factor authentication whatsoever: users are unable to add it to their accounts, while those who had previously set it up on the old site design (including yours truly) are able to log in with just an email and password. At no point in the login process does the site ask you to input a 2FA code.

Humble Bundle does not address the removal of 2FA anywhere on its new site. In fact, the site's FAQs still feature articles encouraging users to make use of it to keep their accounts secure, which is currently impossible. I have reached out to Humble Bundle to ask if the change was intentional and if it will be reversed, and will update this piece if I hear back.

This is, to put it mildly, concerning. The point of two-factor (or multi-factor) authentication is to make user accounts—which, on Humble Bundle, may well have credit card info saved to them—more secure. Even in the event your account password became compromised, your attacker would be unable to get into your account without also having access to your 2FA codes, which are constantly regenerated and only valid for short periods of time. In a world where too many people have bad password hygiene, it's a vital bit of security tech.

Unsurprisingly, Humble Bundle users are aghast at the uncommunicated weakening of their account security. "This is insanely bad, they need to take down password changes entirely until this is fixed", wrote one user in the Humble Bundle subreddit. "I can't even see a way to remove saved payment details anymore, this is seriously dangerous and the site should be taken offline until fixed ASAP", wrote another. They all, uh, they all pretty much go like that. No one is happy about this.

I have to imagine this was not an intentional change. Indeed, I can't fathom the circumstances under which it would be intentional. Either way, while 2FA's absence endures, you might want to play it safe and remove any saved payment details you have tied to your account, though some users report having trouble doing that, too.

In compliance with local laws, Steam rolled out age verification in Australia two months ago. Controversially, it was based on whether you owned a credit card, which many Australians do not. Overnight, Australians stopped being able to see games that are rated R18+ in the Steam store, like Doom, if they didn't have a credit card linked to their account.

Just as suddenly, Steam support has now updated how age verification works for Australians. Adding a credit card to your account still works, but additionally: "Steam will automatically consider your account age verified if it is at least nine years old and has had a consistent email address or billing information."

If you've moved house in the last nine years and let Steam know about it then your billing information won't have been consistent, so I sure hope you haven't changed both your email and physical address in that time.

Valve notes that it is aware this solution still isn't ideal for everyone: "We understand that not all Australians will be able to verify their age using our existing options. We are looking into other ways to comply with Australia's age verification rules while also limiting the data we collect and store from users."

Other storefronts provide alternative options like ID scans and face scans via digital age assurer Yoti. The Microsoft Store also takes your email account into consideration, via VerifyMy, the same assurer YouTube uses.

I really wasn't expecting Microsoft Paint to turn up in the cybersecurity headlines, but here we are. MS disclosed and patched a security vulnerability within the software on October 13, but it's worth checking your own Paint version to see if you've received the update.

Some users don't have automatic Microsoft Store updates enabled, especially those on work machines. As Paint is updated via the store, you may need to manually update to version 11.2609.401.0 to keep your machine secure.

To confuse things further, International Cyber Digest reports that the MS store was providing older versions to some users until recently, and Windows Latest didn't seem to receive the new version even with auto updates enabled. Or, there's been some mixup with release numbers. It's genuinely difficult to tell at this point, with differing information coming from multiple sources.

The vulnerability was initially given an 8.1 severity rating on the CVSS scale, but Microsoft downgraded it to a 7.0 score on November 9 according to Threat Defence. MS also changed its description of the bug to that of a local exploit (requiring a user to open a crafted file) rather than a network attack vector, .

Regardless, to check whether your Paint version is hopefully up-to-date (what a world we live in), open the app, go to the Settings menu, and check the version number beneath the "About Paint" section. If it's lower than version 11.2609.401.0, open the Microsoft Store and manually update the app. Or, if your Windows 11 settings don't allow user updates, contact your sysadmin.

You can also check your Paint version within Windows Powershell by running:

powershell Get-AppxPackage -Name Microsoft.Paint | Select-Object Name, Version

Honestly, this is the biggest omnishambles of conflicting cybersecurity info I've seen in a while. Some outlets initially reported the bug as a no-click security flaw, while others have suggested that the vulnerability could potentially be exploited via a specially crafted image. Microsoft shifting the security goalposts as it goes really hasn't helped, nor has its apparently messy roll out of the patched version.

Anyway, it's still worth checking if Paint has been updated, folks. One way or another, we'll get to the bottom of this eventually. Or we'll just keep updating every aspect of our machines on a tediously regular basis, in the hope that some of these messy leak-plugging antics might do some good.

The Australian STEM Videogame Challenge, a "free national videogame development competition for students," has confirmed that the rise of generative AI means this year's event will be the last.

The end of the STEM (Science, Technology, Engineering, and Mathematics) Videogame Challenge was first reported on Reddit last week, when someone claiming to be a judge in the program said they'd received an email informing them that "after 12 wonderful years," the 2026 event would mark its end.

Speculation as to the cause of the program's cancellation focused primarily on the presence of AI in entries: Another self-proclaimed judge said the problem wasn't merely the use of AI in itself but that "some students just completely generate their games with AI."

"How do you mark that? You can't just give them a zero," the judge wrote. "It's so hard to tell what was or wasn't generated using AI, and especially hard since that's the way the industry is going."

That same person also suggested, however, that funding cutbacks could have been an even bigger factor in the decision, saying that "projects like this are the first to get cut" when government budgets get tight.

Now the Australian Council for Educational Research has confirmed that the cancellation was in fact due to AI.

"After more than a decade inspiring the next generation of young creators, the current 2026 Australian STEM Video Game Challenge will be the competition's final chapter," Cathy Zemaitis, Head of the Student Learning and Progress division at ACER, told PC Gamer.

"The decision was driven by the rapid and complex rise of generative AI in game development, making it harder to judge what students made themselves, and moving the program away from its original purpose."

So, that's that: A program aimed at encouraging kids to learn has come to an end because the kids learned how to get around the learning part. It's sad, although as a former high school reprobate myself I can't be too critical of the young'uns for doing what they do. I do wonder, though, what all of this might say about our future: Where will we be when we know how to prompt, but not how to create?

Zemaitis added that organizers are not looking at ways to adjust the program to accommodate generative AI: "There are no plans to pursue alternative options for delivering the Australian STEM Videogame Challenge, however we're always looking for how we can partner with schools and the sector more broadly to deliver programs that enrich the learning experience of students." She also encouraged educators looking for other ways to "keep students challenged" to check out the International Mathematical Modelling Challenge.

I've spent far too much of my life covering the rampant ascent of AI. A juggernaut of tech development, destined to change the world into some sort of sci-fi utopia (if you believe its advocates) or fated to come crashing down around our ears, taking a large chunk of the world economy with it. Or the end of the world as we know it, according to some.

And now, the pillars upon which the AI revolution sits seem to be shaking beneath it. The most recent tremors began with the wobbling of US AI-led tech stocks in mid-October, after a similar descent in China earlier this year. Then at the end of the month, while many tech companies reported huge amounts of revenue in their quarterly financial results, some AI tech share prices still slid under a black cloud of investor doubt.

None more so than Alphabet, the owner of Google. A $500 billion wipe in stock price, led by the admission that the company's own AI search features were eating into its ad revenue, rocked the market. A wider tech sell-off, fuelled by fears regarding the eventual profitability of AI and a potential upcoming Democrat victory in the US midterms, rattled the valuations of many AI tech companies—alongside the chip manufacturers sitting underneath them.

All this, of course, amid a backdrop of increasing AI-related concerns across multiple fields. Rogue AI agents. Increasing AI-powered cybersecurity incidents. Predictions of the end of civilisation, brought on by rampant AI development without proper oversight. Fears of job market ruination. AI psychosis. The list goes on.

Then, the so-called blue wave came to pass. Democrats won control of both the US House and the Senate. Such sweeping political changes can rarely be ascribed to a single factor, but US voter polling indicates that data center expansion was a key source of voter frustration. The current Republican administration has gone all in on AI, and much of the US public seems displeased at the idea of massive AI tech facilities springing up left, right, and center across the country.

Trillions were wiped off US tech stock values as investors sold off increasing numbers of shares in tech companies, amid further fears that the great AI infrastructure buildout was about to be reigned in by political changes. Long-called-for regulation of AI and AI companies seemed—and indeed, still seems—ever more likely.

And now, Nvidia CEO Jensen Huang has hit the headlines over allegations surrounding his involvement with an alleged Chinese spy. At first, it was reported that Alice Huang (no relation) had a "close relationship" with the Nvidia chief.

Now, Jensen Huang has himself admitted that the two were engaged in an affair, and Nvidia has launched an internal investigation into concerns related to national security. The timing could not be worse for the US tech market. At the time of writing, Nvidia's share value has sunk in pre-market trading.

The company is, in many ways, the backbone of the AI revolution, the chief provider of the chips behind the sweeping AI rollout that has changed the tech landscape over the past few years, and the creator of the CUDA platform that underpins it. Huang is its charismatic leader, the relentless optimist behind this AI-led future we're supposedly about to enjoy. A scandal at this stage is likely the last thing any tech investor wants to see.

So, what now? While the long-mooted AI bubble burst has yet to materialise in full, it's certainly looking like the relentless "all the money in the world" optimism phase is coming to an end. Investors are shaken, with worries about long-term profitability continuing to eat into the market, alongside US national security concerns now dominating the headlines.

The train with no brakes is rocking wildly against the tracks—and unfortunately, given AI and big tech's role in the global economy, we're all stuck in the carriages behind it.

I'm no economist. I can't tell you if this is the beginning of the AI bubble pop, or whether this is merely an extremely turbulent time in the AI success story. Certainly, it seems like the horse has already bolted, and that AI is here to stay in some form or fashion. Those hoping for a return to the good old, pre-AI days are almost certain to be disappointed.

But, having reported on so many "AI doomsayers" over the past few years, I can tell you that a fall from grace has been predicted by many. Such a massive, world-changing technological development, fuelled by seemingly impossible amounts of investment, has always seemed destined to face some sort of financial reckoning.

That reckoning, I believe, has never looked closer. May you live in interesting times, as the old saying goes. And what happens next? Well, it's going to be interesting, at the very least.

GOG Galaxy—the DRM-free storefront's go at a desktop client and launcher—was always a nice dream that didn't measure up in reality. The pitch was: all your games, from all your PC game storefronts, in one centralised hub, but it never left beta in its 2.0 form and, in general, ran pretty poorly compared to its most obvious competition (Steam).

But that was old GOG. New, independent GOG has big plans for Galaxy. In the storefront's first "GOG Showcase", released Thursday, it announced the launcher would be getting a revamp in 2027, and it's coming to Linux too. "We know this has been a long time coming, but we wanted to do it properly: GOG Galaxy is officially heading to Linux", says the studio.

"At the same time, we’ve reworked the client on Windows and Mac to be simpler, run faster, and give you cleaner ways to view your library. Most importantly, our core promise stays untouched: GOG Galaxy is entirely optional. Your games are yours to download and keep through offline installers, exactly as they’ve always been".

That is to say, you'll still be able to find offline game installers in your GOG library if you want to prep for the digital ownership apocalypse (which, let me be clear, is probably a wise move).

Of course, if you're someone who plays games on Linux (cool people, I believe is the technical term), then you likely already have perhaps several tools which do the GOG Galaxy thing of letting you configure and install games from multiple storefronts: programs like Heroic, Faugus, or Lutris. Still, it'd be good to have a ship-shape Galaxy on multiple platforms, and a store with GOG's stature getting in on Linux is good news for PC gaming in general.

The Linuxification of everything isn't the only bit of news GOG released at its showcase. It also announced GOG Social: "a quiet, tidy hangout for actual game talk", meant to centralise discussion and socialising around "screenshot galleries, videos, and community polls with threaded replies and zero algorithmic noise".

If either of those things sounds cool to you, you can sign up for their respective betas over at GOG.

Kenney, an artist known for giving away thousands of free game assets, did not expect to wind up directly arguing with specific billionaires after telling off billionaires in general in a recent X post. But a few hours after he encouraged other artists not to give in to pressure to use generative AI, Epic Games CEO Tim Sweeney was in his replies.

"This is bullshit, Kenney," Sweeney told the artist, concluding his rebuttal with a cute double entendre: "I've upped my expectations of AI. Up yours!"

Richest-person-in-the-world Elon Musk also chimed in to back up Sweeney. Hundreds of other comments on the exchange inevitably followed, spanning the full spectrum of AI opinions.

Kenney's contentious post was: "Don't let billionaires convince you slop is the future. They're used to managing & directing, not making things themselves. Of course they can't see the difference. They don't understand your love for the craft. Keep making personal, beautiful games. You've got my support 💖"

The artist had not had Sweeney in mind when he wrote the post; it was a response to something Musk had said. Kenney expected it to be read by other artists who may feel pressured to use generative AI despite objecting to it.

"I think AI has some benefit but there's just too many problems with it that I find it immoral to use in its current state," Kenney told PC Gamer in a chat after that exchange. "Whether it's the environmental issues, data centers, theft during training, or the negative impact it has on pretty much every industry."

The AI defensive

Sweeney is not known to keep his opinions to himself, but the Epic CEO has been particularly active in his defense of generative AI on X recently.

While some companies use AI but stay quiet about it—Baldur's Gate 3 developer Larian received substantial pushback for discussing limited AI use last year, as an example of how the conversation can go—Epic has made its support for AI tools unambiguous. The company is going "all-in" on AI-assisted game development with its next Unreal Engine release, proudly calls the new version of its Epic Games Store launcher "vibe coded," and recently announced an AI-focused game jam.

Meanwhile, Sweeney has joined likeminded X users in battling what they see as online persecution from an anti-AI "hate mob."

"Because speaking in favor of AI draws such harsh criticism, there is great reluctance among most professional AI adopters in gaming to speak up," the Epic CEO told PC Gamer this week. "As a provider of technology to the whole industry, I see it as Epic's job and my job to make the positive case for it."

According to Sweeney, opposition to AI in games is not a popular view. He has described the anti-AI crowd as "a pessimistic minority running a venomous, vocal pressure campaign to force compliance and silence dissent," and claimed this week that most opposition to AI comes from "astroturfed anti-AI influence campaigns."

PC Gamer has not seen any evidence for that last claim, but it is the case that "AI slop" accusations on social media have not been matched by a massive rejection of AI from the broader gaming audience.

The generative AI use in Arc Raiders led to a 2/5 review in Eurogamer, but the extraction shooter nevertheless sold over 16 million copies. Several games on Steam's current top sellers list also disclose AI use, including Call of Duty: Modern Warfare 4, Aion 4, MapleStory, and indie game Moon Sweepers, which just announced 600,000 sales.

A recent survey of Steam users by GameDiscoverCo found that the majority, about 69%, are either "neutral," "fine with," or "have absolutely no problem with" buying games that disclose the use of AI in their development, something Steam requires for certain applications of AI.

However, the survey found that the disclosures themselves are popular. According to its results, most Steam users check them before buying a game, and over 55% of respondents think Steam should have even stricter AI disclosure rules.

Sweeney is staunchly against Steam's disclosure requirement, calling it "abhorrent" and claiming that it forces developers to make a political statement that incites "cancel campaigns and public struggle sessions" against them.

The anti-AI camp

Kenney's read on the situation is the opposite of Sweeney's: that a minority which includes billionaire tech executives is pushing hard to normalize generative AI tools and get everyone hooked on them despite resistance from a public that wants to see "soul and love put into games."

"If it were true and AI has had mass adoption, why even bother going after me or trying to convince me to start using it?" Kenney asks.

Some gamers do support the use of AI—or, as established, are at least willing to accept the use of generative AI tools—but even as OpenAI boasts over a billion monthly active users, it's easy to find people who share Kenney's total distaste of AI, and the conversation is not limited to mean X posts.

As of the most recent Game Developers Conference survey, over 50% of devs think AI is bad for the industry. Wariness of data centers crosses party lines in the US, as we just saw in the midterm elections. A large survey of PC Gamer readers recently came back strongly against generative AI in games.

One developer went so far as to cancel their Epic Games Store release in support of Kenney. "We just don't wanna be associated with such an embarrassing platform," said the studio, Dead Teapot.

Kenney is a particularly appropriate figure for the anti-AI game development crowd to rally around, embodying the communal, cooperative artmaking spirit that big AI is seen to threaten. For over a decade, Kenney has been releasing free assets for game developers. It started with leftover art from abandoned Flash game projects, and now the artist makes a living from donations, as well as sales of a $20 bundle that packages all 60,000 assets into one download.

"It isn't a lot, which isn't too surprising if you give nearly everything away for free, but it's enough to [get] by," says Kenney. "I'd rather leave the industry someday when I'm old(...er) and have everyone remember me for the good stuff I did and the many people I've helped, rather than for lining my own pockets."

The great impasse

Meanwhile, Sweeney argues that productivity-aids like generative AI are necessary to make game development economically viable today. He recently reposted an AI-generated cartoon depicting a dev's game launch being ruined by an angry anti-AI mob after they used AI to cut their costs, claiming: "This is exactly what's happening."

In the back-and-forth that followed Sweeney's rebuttal to Kenney's "don't let billionaires convince you slop is the future" post, the Epic CEO told Kenney that his reason for pushing back is that "there's risk that convincing devs to abstain from the new way leaves them behind."

"I'm concerned that development studios swept up in these anti-AI campaigns will make decisions that render themselves uncompetitive in the global market for games," Sweeney told PC Gamer this week. He predicts that studios will see "productivity gains of at least 2x to 4x" with AI tools, and says that "all developers need to be prepared for this."

For Kenney and likeminded developers and players, however, a game development scene that's all-in on generative AI would not be a more efficient continuation of the same scene they helped build, but a betrayal of it—something they fundamentally don't want to be a part of.

"I'm not worried about myself or my own studio," says Kenney. "I'm worried that the part of the industry I love is going to disappear or be taken over by people who aren't in it for the love of the craft, but rather for the grift."

And regarding the concern that developers who don't embrace AI will be left behind: "Please, leave me behind," says Kenney. "I'm pretty happy here."

Some implementations of secure shell (SSH), one of Linux's most trusted and secure connection protocols, have been found vulnerable to the most peculiarly simple methods of bypass: spamming characters.

On some distros, simply type 257–1,024 characters and you get access to the root (admin) login. Thankfully, the list of vulnerable distros is quite small and niche, and OpenSSH has already issued a workaround/fix.

In some new release notes, OpenSSH explains that "this release contains a critical security workaround that affects Linux systems that use the musl libc and PAM disabled."

"Sshd(8): avoid password authentication bypass that is exploitable through sshd(8) that affects some Linux systems when built and run with PAM support disabled. We consider the root cause of this bug to be outside sshd, but it is exploitable via sshd and can be avoided there."

SSH itself is a longstanding and ubiquitous networking protocol that anyone who has dabbled even lightly in Linux networking projects will be familiar with. It's what manages comms between Linux machines. If you've ever used a Raspberry Pi remotely, for instance, you'll likely have logged into it using SSH. Same goes for a NAS. It's also used by devs to push/pull/manage GitHub files.

Thankfully, it's not SSH in general that has the problem but just a small number of distros that use SSH.

According to an explanation on sshhole.com—a website written by Ozgun Kultekin who discovered the bug—this is an issue with "musl Linux."

This refers to any Linux OS that uses the musl C library rather than the standard Glib C library, which is a fancier way of saying: not many distros.

Most distros don't use musl, but a few do. Perhaps most prominently, Alpine Linux, a minimal, security-focused distro. As OpenSSH clarifies, it's also not just musl Linux distros in general, but only those running SSH servers that have Pluggable Authentication Module (PAM) support disabled.

The background as to why this bug occurs and why it was overlooked is pretty complicated, involving different login security functions developed for different purposes overlapping with this specific kind of edge case being overlooked. Sshhole.com explains it better than I can.

The long and short, though, is that one function used * (a star) as a value under the assumption that it would never be used as a password hash, because that usually cannot happen. But another completely separate function, after some further shenanigans, makes the password hash a star if it's 'invalid' because the password has too many characters.

Thus: Spam letters, get let into root.

Most Linux users are absolutely fine. As the bug researchers explains: "On debian, ubuntu, fedora or anything else with glibc? going through pam? dropbear? openbsd? you're fine."

Here are the set of conditions you need to fall under to not be fine:

• "Is your sshd running on musl? (alpine, openwrt, buildroot, yocto)
• is it not using pam? alpine's openssh-server doesn't by default, or maybe you set UsePAM no
• are password logins still on? they are by default
• is there an account with just * as its password? key-only users, ldap users, root on alpine
• and some other account with a real password?"

Though for my part, and although this might not be warranted, I'll be considering this will a justification for sticking to the bigger distros. Ubuntu, Fedora, and so on. Or Windows, I suppose. Good things are still happening over in closed-source land.

Open up Task Manager and check how much memory Discord is currently hogging on your PC. If it's anything like mine, it's probably north of 500 MB. Alas, things should change soon, as Discord is moving the client completely over to a new programming language. It might not sound like that big of a deal, but Discord's CTO, Stanislav Vishnevskiy, shows it running on as little as 290 MB of memory.

Discord is switching from Electron to Rust. Electron is a software framework, a building block that allows JavaScript apps (among others) to run standalone on desktops. Built on Chromium, the underlying model behind Google Chrome, there is an overhead here that swapping to Rust aims to fix.

Vishnevskiy says, "We've been early adopters of Rust since 2018 and a lot of our infra, voice tech, and even some of the current client is already Rust, the holy grail was getting the whole client there. That means Electron goes away, and so does React Native, mobile keeps native UI but shares app code with desktop via Rust."

Though some suspected Discord would be swapping to Rust using Tauri (an open-source framework for desktop and mobile platforms), Vishnevskiy says it's actually using GPUI (a framework written in Rust), but "We have our own patches on top of it".

Vishnevskiy also says the goal is to get the program into testing with users soon, but no official date has been given. Either way, with Vishnevskiy showing off a resource demand of 290.2 MB whilst active, that's still a pretty big dip below what it runs at now. Mine is currently flitting between 500 and 600 MB, and my colleague Jacob saw it go up to 785 MB running on Windows.

Reportedly, the new Discord client is also faster to use, booting much quicker. Vishnevskiy showed two laptops, one with the old client and one with the new, and the new client opens up significantly faster (six seconds versus one in my approximation).

Naturally, we are hearing this from the mouth of the company, so we'll have to get the new Discord client in our hands to see how much faster and how much more efficient it really is, but anyway, to get a little more out of your current memory will always be a boon right now.

The memory crisis is continuing to drive the prices of gaming gear up, but software tweaks like this (including both Epic and Steam getting faster) are very much welcome in my eyes.

I've spent all night playing The Legend of Zelda: Ocarina of Time remake on the Switch 2, and it's a fantastic experience, but some would prefer to tinker and toil over an emulator than enjoy Link's new look. The game is already running on PC, but performance issues are aplenty.

You can see progress is moving pretty fast from the emulator crowd, just from these two videos by Unreal Gaming PC on YouTube.

Published on the game's official launch day, November 5, this video shows the game getting over the splash screen and running the cutscenes fairly well; however, it suffers from bad flickering, stuttering, and frame rates around the 20-30 fps mark during gameplay.

12 hours later, another video. This time showing the game running between 30–60 fps during gameplay, albeit with some persistent flickering and stutters throughout. There's undoubtedly still a long way to go to a playable experience to match the native Switch 2, but that's marked progress for half a day's work. Especially considering the game is designed for an Arm-powered handheld with very different hardware to what's running the emulator (Ryzen 7 9800X3D, 64 GB of RAM, and RTX 4080).

User akakevinvro on X also posted a video claiming a pretty steady 60 fps.

Both are running on the Lichii emulator for the Nintendo Switch and Nintendo Switch 2. It's called the "VERY experimental Nintendo Switch 2 emulator" on its GitHub page. It's an open-source project that has similarities to projects of the same ilk, such as Yuzu. Yuzu's developers settled a legal dispute with Nintendo of America and shut down development of the emulator in 2024.

Nintendo has a long history of taking down projects it deems to infringe on its rights, while emulators such as the now defunct suyu often argue that they do not circumvent Nintendo's privacy measures.

In this case, the files for Ocarina of Time's remake were reportedly dumped online ahead of its release, according to SrLinguicaVIVE on X.

Xbox has joined the 'faster launcher load times' arms race, announcing that its updated Xbox PC app, available now to Xbox Insiders, launches up to 60% faster, and has also been overhauled to offer "simpler navigation, updated filters, and new ways to organize your games."

Along with the faster launch time, Xbox said the home, store, and Game Pass pages in the updated app load up to 50% faster, while in-app windows open up to 65% faster—depending upon factors including device, internet connection, system configuration, and usage, of course. Library navigation has been simplified, search has been improved, and friends and social features are now part of the main navigational window.

Credit where it's due: That change to friends and social features is welcome. I don't use the Xbox App much—to be blunt, it's pretty much exclusively for Sea of Thieves. But after a long stretch away from it, I recently jumped back in for some high seas adventures with the crew, and I'll tell you this: Figuring out how to connect with my fellow online swashbucklers was kind of a headache compared to the super-simplicity of Steam. It worked fine, it just wasn't smooth and easy, so anything that helps with that, I'm happy to take.

I'm less excited about the faster loading times. Not that I'm going to complain about performance improvements, I'm just not sure it matters in the grand scheme of things. When the Epic Games Store promised that its "vibe coded" launcher was faster than Steam, a common reaction was mockery: Much of that was sparked by Epic's insistent banging of the 'AI is inevitable' drum, but there was also a lot of 'who cares?' in the reactions, too. I think the Xbox PC app may face the same: Faster is good, sure, but is it going to convince many PC gamers to use it instead of Steam when they don't have to? Of this, I have doubts.

(Steam also recently dropped a performance update that promises to improve loading times as well, blunting the impact of both the Epic and Xbox boosts even further.)

If you want to check out the updated app for yourself, you'll need to be in the Xbox Insiders program—for that, you'll need to install the Xbox Insider Hub. A "refreshed Xbox web experience" is also on the way, with an improved library tab, more efficient use of screen space, and light, dark, and high-contrast visual modes.

OpenAI CEO Sam Altman said on a recent podcast that "from a global allocation perspective, actually it's quite terrible to have everyone have a bunch of the world's limited supply of DRAM sitting in their own computer, and be used 5% of the time."

Altman was talking about people running AI models locally rather than relying on data centers, which allow RAM usage to be maxed out in this "very supply-constrained environment." What many heard and responded to—PC gamers in particular—was resentment directed at those of us who don't run AI companies, but would still like to have and use our own personal computers, as a treat.

In an exchange on X this week, Altman said that his remark was "taken out of context" and that he actually wants "the world to have tons of memory and people to put as much as they want in their computers."

"I have been arguing for the world to ramp up memory production for a long time," Altman said.

Altman's argument for increased memory production is no consolation to the tech enthusiasts and gamers who've been priced out of new hardware by his company and the rest of the AI industry. What we've been calling the RAMpocalypse hasn't just affected high-end PCs, either: In the lead-up to GTA 6's console launch, a PlayStation Pro is $900, if you can even find one.

The bitter reaction to Altman's comment also comes from concerns that big tech companies want to push gamers off of local hardware so that they can rent us cloud computing. For an audience that's already disgruntled about the erosion of game ownership, the thought of losing control over our own hardware is not palatable, however much cloud streaming services like Nvidia's GeForce Now have improved.

That concern is now coupled with the generative AI-heavy future that tech leaders are pushing. If generative AI is best delivered as a cloud service, and the games industry is going to increasingly incorporate generative AI into games, it follows that we will be pushed toward cloud computing.

Nvidia DLSS pioneer Bryan Catanzaro made a comment to that effect at a GDC panel I attended earlier this year. Speaking about hypothetical games built with AI "world models," Catanzaro noted that we probably won't run those models locally, because "AI is fundamentally much more efficient in the cloud."

Some air was let out of the AI bubble this morning as tech stocks took a dive. After seeing gains on Wednesday, the S&P 500—the index which tracks big companies including Nvidia, Oracle, and other bettors on the AI buildout—has dropped by around 3.9% as of 2:30 pm EST.

Nvidia shares are now trading lower than they were at the start of the year. Memory and storage company Micron is down about 15%.

Democratic victories in the US midterms, and country-wide votes to curb data center expansion, have contributed to the reduced confidence in AI stocks.

Speaking to CNBC, One Point BFG Wealth Partners chief investment officer Peter Boockvar said that investors are now "questioning" the $1.2 trillion in AI-related capital expenditure Wall Street has forecast for next year. "Well, maybe that's not going to happen," he said.

A chain of events that led to a record-breaking $6.3 billion mega-hack began with a dodgy download, so says the company that was hacked, Binance.

The world's largest cryptocurrency exchange says that the hack was not a "single isolated breach or simple phishing incident"; multiple steps were required to break through its security measures. It all began when a Wallet DevOps engineer (named Engineer A) downloaded a tampered Homebrew installation that was suggested to them by an unnamed search engine.

Homebrew is a third-party package manager for macOS. A report from the SANS Technology Institute, linked to in Binance's investigation, suggests that malicious adverts for Homebrew have been spotted in search engine results, hoping to ensnare users into installing malware on a user's PC.

"Our security team detected the anomaly, replaced the device, and rotated all related credentials as a proactive security action," Binance says. "However, Engineer A’s role and his related colleagues on our wallet infrastructure team was exposed to the hacker."

Then came a separate incident with a software engineer (named Engineer B). They were contacted through a job hunting platform (Binance says it was a legitimate platform) and was asked to attend a first-round interview with a HR representative. This was followed by up a second-round interview with the CEO. Not the HR representative, nor the CEO, however, were real; the company was made up.

The aim of the interviews? To offer Engineer B an 'assignment' that was really a repository of malicious code that executed during setup. The malware ran a series of commands and evasion techniques to avoid detection, thus avoiding Binance's security checks.

The attacker used Engineer B's now-compromised computer and an internal AI tool to gain control over two further machines: that of Engineer A and another Wallet Software engineer (Engineer C). They discovered a container deployment permission token on Engineer A's machine. With this, they could log into Engineer C's computer using the stolen credentials to create a container and trigger a malicious script to start the mega-hack that would drain $6.3 billion worth of funds from wallets used to store transitory funds by the exchange.

So, basically, the internet is a bad place, and you can't trust anyone. Or, at least, you have to have your wits about you—especially if you happen to have the key to a multi-billion dollar company's finances stored on your PC somewhere. Luckily, my PC is just pictures of my cat and a Steam backlog that won't quit.

That's actually some of the criticism levelled at the company. Should it have been possible to get this level of access even with spoofed ads and jobs?

And here's Binance co-founder Yi He on taking that criticism...

Reminder that the FBI reckons a North Korean hacking group is behind the attack, which North Korean officials have unsurprisingly denied. Binance has frozen roughly half of the funds that were initially stolen, and has already covered the rest.

Saying that it believes "schools should have control over AI in their classrooms," Google announced today that it will turn off access to its AI tools for K-12 student accounts. But there's a catch: Along with those AI tools, it will also disable Google search functionality.

In a blog post entitled "Designing responsible AI for learning," Chris Phillips, Google's vice president and general manager of education, extolled the virtues of AI tools in "productive struggle," which he described as "enabling students to be actively challenged without getting so overwhelmed that they give up."

"That's the philosophy behind Guided Learning," Philips wrote. "In Gemini, Guided Learning acts as an interactive study tool that coaches students through the process of problem solving through open-ended questions and step-by-step breakdowns. Our research shows that Guided Learning leads to measurable learning gains for students, and we're now bringing Guided Learning to more products." Those products include Google Search and Google Classroom.

"After listening to schools, teachers, leading researchers, and policymakers, we are now updating our approach to controls for Gemini, Gemini Notebook, Gemini in Classroom, and Google Search," Phillips continued. "Access to these products will be entirely off for K-12 student accounts until school administrators explicitly turn them on."

It sounds like Google Search on student accounts can be re-enabled separately, but that also ushers AI back into the equation: Google said in the update that it is "building Guided Learning capabilities into Search for student accounts to help them work through problems without skipping straight to answers." In other words, there is no search functionality on these accounts without AI intrusion.

Speaking to the New York Times, Phillips said "search and AI are deeply integrated," and confirmed that schools can't have one without the other: Students at schools that have turned off Google's AI tools "will not have access to search, and they'll see a message from their school that indicates it's not available."

Phillips presents the policy as responsiveness and responsibility on Google's part, but the reaction from educators is not entirely positive. American Federation of Teachers president Randi Weingarten told NBC that "Google has prioritized its business interests over the education of our children," adding, "Instead of working with teachers and parents to uphold privacy and ensure their tools are safe and useful, it throws the baby out with the bathwater and bans access to the tools altogether."

Fairplay, a non-profit that aims to protect children from "the exploitative and harmful business practices of marketers and Big Tech," was somewhat harsher in its assessment. "Google's attempt to strong-arm schools into using its AI products is likely to backfire," Fairplay executive director Josh Golin said in a statement. "Schools should use this opportunity not only to turn off AI, but to teach students that 'Google' isn't a verb, but the name of a surveillance and advertising company whose presence in schools has nothing to do with learning and everything to do with developing lifetime brand loyalty.

"Because AI isn't safe for kids and undermines education and critical thinking, schools should turn off Google's AI tools the minute that option becomes available. And rather than fret about losing access to Google search, schools should use this opportunity to help students find more privacy-protective alternatives for search, as well as ways to do more substantive research beyond search and click."

Former FTC chair Lina Khan, who was right about Microsoft's acquisition of Activision Blizzard, also weighed in, accusing Google of "using its search monopoly to coerce schools into adopting AI."

"Multiple courts have found Google liable for violating antitrust laws, and this latest move shows it remains unchastened," Khan wrote on X. "As the public increasingly pushes back against the use of AI in schools, Google is making clear it will punish those who don't fall in line."

New York City mayor Zohran Mamdani, a longstanding critic of oligopolies and corporate greed, said that "Google is trying to use its monopoly power to force AI into our public schools."

Google is trying to use its monopoly power to force AI into our public schools.No corporation should have the power to dictate how we teach our children. Our public schools exist to serve students, not to grow the profit margins of Big Tech.

— @mayor.nyc.gov (@mayor.nyc.gov.bsky.social) 2026-11-03T22:58:02.432Z

As NBC noted, research on the impact of AI in education is still quite limited, and some studies purporting to show meaningful benefits have been found to be flawed. That makes the headlong rush into AI integration, championed by massive AI companies and pretty much no one else, so controversial and potentially dangerous. Trying to force AI into classrooms while the technology's impact on mental health is an increasingly serious concern is audacious, and promises of more rigorous safety standards tend to ring hollow anyway: In October, the attorneys general for eight US states accused competing product OpenAI of ignoring its own safety warnings in pursuit of profit.

Google's new policy on AI in schools comes just days after shares in Google parent company Alphabet took a hammering as the company's AI Mode cut deep into its search business and revenue. I've reached out to Google for more information about the new educational policy and will update if I receive a reply.

Indie-oriented digital storefront Itch.io is beginning to push back against the rising flood of AI, announcing on social media that it's "deleted a couple terabytes of AI slop" and made it easier for users to filter out games made with generative AI.

A few days after Valve admitted that Steam has a problem with AI slop and game cloning, Itch.io chimed in to say it's experiencing something very similar: The number of game pages being created every day on the storefront has doubled over the past two months.

Itch.io didn't provide a reason for the phenomenon, but it's not hard to figure out: When one Bluesky asked why it was happening, Itch.io replied "take a guess," accompanied by crying and robot smilies, and then added that "the number of downloads and purchases has *not* doubled."

Now it seems to be moving to action. "Just deleted a couple terabytes of AI slop," the storefront posted last night. "Just because you can upload 10 terabytes of poorly optimized AI generated models doesn't mean you should upload 10 terabytes of poorly optimized AI generated models."

The account also linked to Itch.io's guidelines on "uploading excessive amounts of automatically generated or AI-generated content," which says that while content created by generative AI "can be useful or interesting, do not use automated systems to mass produce product pages. We will consider this spam and may remove your pages from our browse and search pages."

The guidelines apply to, but are not limited to:

• Games or projects that are predominantly created by algorithms or AI with minimal human intervention
• Content that is randomly generated with no or minimal curated design
• Multiple iterations of essentially the same content that are generated by AI or algorithms

Like Steam, Itch.io requires developers and publishers to disclose their use of generative AI in their games. That disclosure requirement has now been expanded from game assets to "the tabletop (aka physical games), books, comics and soundtracks categories," all of which now require disclosure.

It has also added new browsing and search filters on the site, so users can exclude some or all use of generative AI.

The tabletop (aka physical games), books, comics and soundtracks categories on itchio now require completing an AI disclosure in order to edit or create your page. These join the game assets section which has required disclosure for the past 2 years.

— @itch.io (@itch.io.bsky.social) 2026-11-03T19:42:09.935Z

The "No AI" tag page cuts straight to it:

Itch.io's action isn't likely to stem the tide of AI slop that's engulfing the world of gaming. Apart from the fact that it represents a very small percentage of a market dominated by Steam, the simple truth is that AI slop is easy to churn out in overwhelming quantities—that's what makes it so attractive to people who care less about creation than they do about making a few bucks on a lucky swing of the bat.

But it is action, and I think that's important: While some argue that mandatory AI disclosures have the potential to harm developers, the gen AI geyser is unquestionably doing harm right now, to users, developers, and the industry as a whole. Unless and until that changes, proactively excising the worst of the slop torrent and making it easier for gamers to avoid the mess entirely is the least that online storefronts can do.

Gaming on Mac has come a long way. Gone are the days of limited releases, though not absolutely everything is playable on the platform. Thankfully, community developers are only too keen to plug the gaps.

FX Steam Launcher on GitHub brings SteamOS to Apple Silicon Macs. Specifically, this independent, open-source project is based on the ARM64 Steam Frame version of the operating system. Therefore, you're getting Windows game compatibility on Mac thanks to Proton 11 and FEX, alongside the very swish SteamOS interface (via Videocardz).

Obviously, there are a few things to bear in mind if you want to go all in as a Mac gamer. For one thing, this is not a native installation, and is instead running SteamOS by way of a virtual machine on your Mac. For another, this is a GitHub download to begin with, but once you've got it installed, the launcher directly downloads the official SteamOS image from Valve and then even handles OS updates too. Neat!

As this is an unofficial effort, there are a few more caveats still. The FX Steam Launcher itself currently only supports Apple Silicon Macs running either macOS 15 (from September, 2024) or newer. However, to actually get the latest and greatest running, you may need something much more recent than that.

Bear with me here, but Proton and FEX are essentially only the first hurdle. DXVK is required to run DirectX 9, 10, and 11 games, while DirectX 12 releases rely on VKD3D-Proton. Then either MoltenVK or KosmicKrisp is the next hurdle graphics instructions will need to pass before finally handing off the baton to Apple’s Metal API (though there's some experimental support for DXMT, which allows DirectX 10 and 11 games to run without VKD3D-Proton).

Anyway, all of that is to say KosmicKrisp, a pretty pivotal translation layer in this whole process, requires at least macOS 26 or later to run. That's a fair amount of hoops to jump through, though; if you can get it running, the launcher does allow you to run SteamOS either windowed or fullscreen, while letting you still access macOS either way.

On account of not being a Mac power user, I've not had a chance to test it out myself, but YouTube creator ETA Prime tested the launcher with The Witcher 3, Stellar Blade, and Left 4 Dead 2 on their Apple M4 Max. At high settings, the creator saw performance to the tune of over 70 FPS for each of these games, with Left 4 Dead 2 breezing past 200.

That said, Stellar Blade was running at a mere 800p resolution, while the rest of their tests ran at a maximum of 1080p. Furthermore, not all the games they tested even ran as well as these cherry-picked examples (to say nothing of The Witcher 3's funky-looking trees).

Still, what's possible with this launcher remains impressive, though gaming on Mac even without it is far more viable than it once was. For Apple's big 50th birthday earlier this year, our Jacob explored the things Mac does better than PC. Our Ian was also mighty impressed with the Apple Mac Studio M4 Max—though he does also note, "Christ, what are you doing? Buy something else for gaming".

On that front, any would-be Mac gamer could do much worse than the MacBook Neo, which is now going for $699/£699. Not even Apple is safe from the memory supply crisis, but as the situation continues to drive up the cost of traditional gaming PC parts, who could blame a gamer on a budget still picking up the MacBook Neo instead? Especially with unofficial projects like FX Steam Launcher at their fingertips.

Binance saw a huge outpouring of withdrawals when it re-opened to customers on Sunday following a hack for the history books. The cryptocurrency exchange lost $6.3 billion, roughly $3 billion of which was frozen in transit and the rest covered by the company, but it lost a bunch more once it let customers take their money out on November 1.

Customers withdrew $4.649 billion from the exchange in just under 10 hours since it reopened. That's a whole lot of value to lose in a day, and not wholly surprising given the context, but what is more surprising is how many deposits were made in that same timeframe: roughly $1 billion worth.

"Withdrawals were temporarily suspended as a precaution while our teams worked to contain the incident loss, assess its impact, restore affected services, and complete security checks," Binance said of the move.

Users may face slowdowns as systems get back up and running, the company has said.

In later updates, Binance says that deposits into the exchange have begun to creep back up in excess of withdrawals; the total cashflow of the crypto exchange is still a net negative, at -$5 billion by November 3, but the withdrawals are no longer outpacing the deposits by any significant amount. It's still early days and a long way to go to make up for the initial loss in customer funds, but the exchange may still live to fight another day.

Meanwhile, the exact details of the attack are still being deliberated. The FBI holds North Korea responsible, agreeing with experts that a group going by the name of TraderTraitor is behind it. The FBI says that it, along with USSS and IRS-CI, has observed the group converting the stolen crypto into Ether, with a view to launder it away into obscurity later.

North Korea, unsurprisingly, denies the FBI's claim. Kim Yo Jong, an official in the NK government, and Kim Jong Un's sister, said it was nothing more than a "sinister attempt" by the US to damage the country's reputation.

"The US has made a trite provocation of linking it with the Democratic People's Republic of Korea again," Kim Yo Jong said.

Valve recently admitted that it has a problem with AI slop games and clones flooding Steam—it doesn't know what to do about it just yet, but it at least acknowledges that it's an issue. Now we have Roblox, speaking from the other side of that coin in its Q3 2026 shareholder letter, in which it bragged that the July release of Build, its AI-powered "prompt-to-game creation tool," users have published more than 70,000 games.

"During the last three months, we have witnessed an inflection point in AI adoption and content generation in the gaming industry overall, and on Roblox in particular," the letter states. "AI-enabled game creation, automation in code generation and AI supported asset creation are leading to an explosion of content. The time it takes to go from an idea to a playable game is compressing dramatically, and so is the time it takes to adapt a game for new devices."

The "explosion of content" remark is what jumps out at me first and foremost. Is that really a good thing—content, in explosive quantities? That is essentially what Valve is facing on Steam that's widely considered a problem—an ever-growing mountain of content—yet in the world of Roblox, it's a feature, not a bug. And Roblox is not looking to stem the tide.

"We have been embracing AI to accelerate creation since the first release of Roblox Studio Assistant in December 2023," the letter continues. "Over the past month, more than 80% of the top 100 games on Roblox (ranked by hours engaged) used AI tools to accelerate creation, and the number of daily asset uploads to the Roblox cloud tripled month-over-month.

"Since Q2, time spent using AI tools (including Build, Studio Assistant, and external AI via MCP) tripled among our top creators, and those using our MCP server daily grew nearly 5x. In Q3, Break and Steal an Egg and Huss Valley were both created almost entirely with AI and both broke into the Roblox top 20 (ranked by concurrent players). Build is our new prompt-to-game creation tool. Since its alpha launch in New Zealand, Singapore, Serbia and Thailand, more than 500,000 users have used Build—and have published over 70,000 games."

Take a moment to think about that: Build launched in July, five months ago, which means we're looking at a rough average of 14,000 games per month. And that's with Build still not fully launched yet, so only a fairly small portion of the Roblox userbase has access to it. I can't help but wonder, with a certain sense of alarm, what will happen when the doors are opened to more populous parts of the world.

Roblox, however, says it is "uniquely positioned to capture this moment," which is seeing "traditional barriers to software and content creation" giving way to a focus on "distribution, audience aggregation, monetization, discovery, and safety."

To further capitalize on that, Roblox has kicked off two new initiatives: Roblox Everywhere, revealed in September as a way to make Roblox games playable outside of the Roblox platform, and the Universal Player, which is basically the reverse of that: "AI is used to seamlessly import any existing game into the Roblox cloud platform while transforming it to run multiplayer, on any device (at the upper edge of the device's performance capabilities), with instant join, instant UGC updates, local 3D, and access to Roblox services for discovery, safety, communication, and monetization."

That's interesting: Details are scant but the idea seems to be to make Roblox a platform for all game developers, akin to Steam, GOG, or the PlayStation Store, with minimal fuss or muss: Just make your game, then zippity-pow it through the AI machine, and suddenly you've got a full-on Roblox game and access to a massive audience that previously would've been much more of a headache to reach. And Roblox gets basically the same: Its goal "is to capture 10% of the global gaming market and an even greater share in the US."

"As content becomes abundant, attention and connection become scarce," the letter says, extolling the virtues of Roblox's platform. "What sets our distribution apart is that it runs on social connections: people come to Roblox to play with their friends, so every new game can reach players through an existing social graph, through friends inviting friends, and through social influencers, which we believe enables great content to find its ideal audience faster than on any other platform."

Maybe Roblox is right. Hell, maybe EA chief Andrew Wilson was right—maybe the future really is all about billions of people cranking out content while corporations fight over where to put it all. It's hard to argue that Roblox hasn't put up serious shareholder value so far, after all, and while Steam doesn't have shareholders it's still making bank, AI slop problem or no. Still, I can't say that making speed and quantity the key indicators of success really represents the kind of future for videogaming that I'd like.

More and more, AI appears to be the reason we can't have nice things on the internet. Beyond generative models gobbling up my own and my fellow creatives' work in a bid to produce abysmally average output, AI data centres are also devouring all of the available memory. This drives up prices and makes DRAM supply for us PC gamers pretty scarce. CEO of OpenAI Sam Altman argues this is truly for the best.

Appearing on a recent episode of the Cheeky Pint podcast, Altman argued, "From a global allocation perspective, actually it’s quite terrible to have everyone have a bunch of the world’s limited supply of DRAM sitting in their own computer, and be used 5% of the time."

Yes, that is steam escaping from my ears. But before I go full screaming tea kettle, let's take a beat. Altman is talking less about PC gamers specifically here, and more about solo AI bros craving compute power and "local inference".

He's bemoaning an inefficiency within the AI industry rather than taking aim at us pesky gamers. He goes on to add, "You want maximum utilization in the cloud if you're trying to deal with this very supply constrained environment. And somehow everyone's... Many people's intuition is the opposite of that."

It still feels a bit rich for Altman to complain about the memory scarcity OpenAI and ChatGPT helped create. I could also keyboard warrior my way through several paragraphs about how resource hungry the data centres he's arguing should enjoy DRAM priority are, not to mention how the rapid AI infrastructure build out has caused many of big tech's greener energy goals to fall by the wayside—but I'll spare you. Our planet's environmental predicament doesn't begin or end with Sam Altman or OpenAI, after all.

Instead, I'm choosing to extract a grim sense of satisfaction from the AI snake eating its own tail. For one thing, Google's AI features appear to be cannibalising ad revenue from search. For another, Amazon's big data centre build out has made free cash flow negative, and for just one more thing, investors haven't been overjoyed with the recent performance of many AI stocks, leading to a big sell-off earlier this month.

The bubble may not have burst yet, but the aforementioned snake is looking a little green around the scales.

Maybe I've been too harsh on those mandatory cybersecurity training courses that my company keeps reminding me to do. What looks to be the largest ever cryptocurrency heist may have started with an engineer's compromised laptop, and the attacker got away with more than a couple coins: $6.3 billion in stolen cryptocurrency.

The victim is Binance, by far the largest cryptocurrency exchange in the world. The hacker made off with tremendous sums of a range of cryptocurrencies, including ethereum, bitcoin, solana, and XRP. The total sums amount to around $6.3 billion worth.

Binance says it has worked with blockchain partners to freeze roughly $3 billion worth of that, leaving $3.3 billion still unaccounted for.

Binance founder Changpeng Zhao took to X to suggest that he feels all user funds would be fully covered. Zhao is no longer running Binance following a four-month stint in prison for violating anti-money-laundering laws. He was later pardoned by president Donald Trump. Binance has since confirmed that, yes, it will cover all losses.

"Binance's initial investigation suggests that the security incident may involve a compromised DevOps engineer’s laptop," the company says on X, "which allowed an attacker to gain access to a service container and submit unauthorized withdrawal transactions."

Binance says it's still figuring out what actually happened here, though experts are already chiming in with what might have gone down.

"The attackers didn't break the encryption or steal the private keys. They took over the system that tells the keys what to sign," Jayant Dave from Check Point tells BBC News.

Also speaking to BBC News, Emily Taylor, co-founder of the Global Signal Exchange, said the hack was conducted "on a speed and scale we've never seen."

Blockchain intelligence firm TRM says that the theft may have come down to misuse of what's known as a hot wallet, "a wallet connected to the internet that an exchange uses to process customer withdrawals". Early reports suggested someone had phished the authorisations required to access one of Binance's main Ethereum hot wallets, but later reports of further thefts from other cryptocurrency wallets with different chains, keys, and signing schemes suggest the attacker may have had deeper access to Binance.

"The most likely explanation is that the attacker gained the ability to make Binance's hot wallet signing system sign transactions of its choosing across chains," TRM says.

"In this incident, a single actor obtained valid signatures for wallets on ten chains within hours."

So whodunit? TRM points a finger at North Korean hacking group TraderTraitor, a state-sponsored group said to be behind the $1.5 billion Bybit hack in February, 2025.

Binance says it is working with law enforcement to confirm the identity of the hackers. It hasn't given any official indication as to their identity as of yet.

There are sure to be a few reckonings for the cryptocurrency market out of this. Firstly, how could a single compromised laptop issue $6.3 billion worth of dodgy signatures? Secondly, if cryptocurrency is so decentralised, how come a few key players freeze billions in assets when required? Thirdly, how many large-scale heists have to take place before trust in the system falls through the floor?

Reddit will become a "logged-in only experience," and the company plans to move fast, beginning testing soon in regions outside the US before flipping the switch globally.

The popular community website—home to a great many official and unofficial gaming message boards—lost one million search engine referrals per week on average in October, CEO Steve Huffman said on an investor call today.

The culprit, of course, was AI. A Google search that once might have led someone to a 14-year-old Reddit thread containing the answer to a rare household cleaning problem is now likely to present that answer in an AI summary, no click required.

"Google has really shifted in their relationship with the rest of the internet," Huffman said during a Q&A.

The CEO emphasized, however, that most of Reddit's revenue (over 80%) comes from logged-in and app users, and said that the company will focus on establishing the site as "a direct daily habit" that doesn't rely on search traffic.

Efforts cited in that direction include algorithmic feeds, improved app design and performance, AI-powered moderation tools, and "new experiences like games and visual podcasts," the latter referring to, I believe, a feature that uses generative AI to turn posts into videos.

The switch to a logged-in only experience will also help "protect access to Reddit's extremely valuable corpus of data," Huffman said.

Reddit currently has AI training arrangements with OpenAI and Google. In the Q&A portion of the call, Huffman said that Reddit has nothing to announce about those deals, but that it is in "active discussions" and that its leadership will "do what makes sense and is in the best interest for Reddit strategically and economically."

Although it reported 53% year-over-year revenue growth in its 2026 third-quarter results, as well as 16% growth in weekly active users, Reddit's share price dropped today due to a lower-than-expected forecast for the next quarter.

Jensen Huang, Nvidia's CEO and one of AI's most important people, was reportedly targeted by a data centre executive working as a spy for the Chinese government, a new report suggests.

According to the New York Times, citing undisclosed sources that were briefed on the matter, Alice Huang, also known as Huang Le, was being watched by US agencies due to suspicion of involvement in data centre chip smuggling from Singapore. During that surveillance, the officials are said to have determined that Alice Huang was acting as an intelligence asset.

Moreover, those officials are also said to have determined that Alice Huang had a close relationship with Jensen Huang (no relation). The NYT suggests that officials were concerned that Alice Huang may have been able to extract information from this relationship that would be useful to the Chinese state—what with Nvidia being pivotal to America's AI efforts.

It's said that Jensen Huang had, at the time, not been briefed on the threat of foreign intelligence services. It was later implied Nvidia was not briefed on Alice Huang, either. The Nvidia CEO has a close relationship with US president Donald Trump, and has travelled with the US president to China and was more recently called by the president while on stage at an AI event.

The US has approved the sale of some AI chips to China, though they are not the cutting-edge Blackwell chips that US firms have access to—providing they pay a pretty penny, anyways. China had been blocking some of these chips on its side, but recent reports suggest older Nvidia H200 cards are reaching China in small amounts.

As it stands, the White House, CIA, and NSA have all declined to comment to the NYT. Who knew what, and when, is still up in the air. Even Alice Huang's status with the data centre company that was under scrutiny for chip smuggling, Megaspeed, is not currently clear.

Nvidia has also pushed back on a few of the claims made in the NYT report. The company says that claims Megaspeed had any involvement in chip smuggling were debunked last year in a statement to CNBC.

"The diversion allegations against this regional cloud were thoroughly investigated and debunked last year,” says spokesperson John Rizzo.

Rizzo also pushed back on Huang's relationship with Alice Huang being anything exceptional to the NYT.

“Jensen’s interactions with Ms. Huang are not unusual or unexpected. Meetings with customers are part of a CEO’s job; Jensen talks with thousands of customer executives every year.”

He also added that Nvidia believes the US government would have made it known had there been any threat to American industry, and that Megaspeed would have received "ordinary business information" and not trade secrets; "zero information relevant to national security."

Even during a good financial year, a big tech investor is seemingly never satisfied. Case in point, while Google Search grew by 12% year-over-year and made $63.3 billion this past quarter, parent company Alphabet's share price took a pretty spectacular tumble.

Despite making an amount of money the mere thought of which sucks the air right out of my lungs, Search's revenue this quarter was actually below Wall Street's expectations. As a result, Alphabet's shares are down 13% and $500 billion has been wiped from its market value (via The Financial Times).

The share price is continuing to fluctuate and, according to CNBC, it could be the worst post-earnings call decline ever. This follows $1 trillion worth of air escaping from the AI party balloon earlier this month, due to anxious investors widely selling off stock. So, how did Alphabet get its ABC's so wrong? Long story short, Google's own AI features may be eating into its Search revenue.

This is despite Search apparently seeing the fastest year-over-year query growth rate in half a decade. During Alphabet's most recent earnings call, Google's senior vice president and chief business officer Philipp Schindler was asked to explain this 'disconnect'.

"It's worth noting that the strong growth in AI Mode queries had an impact because of the lower monetization we were seeing in AI Mode," Schindler says. "To be clear, we feel very good about our trajectory here. Our ad experiments have shown promising results, and we're now scaling ads within both AI Overviews and AI Mode to more users and more queries, including, as I mentioned, international markets and multiturn ads in the US."

In other words, yes, AI Mode is eating Search's lunch but Google is hoping to scale up its kitchen production in order to keep both of its babies fed. Or something like that—the metaphor got away from me in the middle there.

Anyway, these financial results only serve to add fuel to the fire of investor fear that Google Search is getting singed by the likes of ChatGPT. I mean, compare and contrast the amount of ads you see on the first page of Google results against what you typically see from an LLM chatbot. As Google tells it, more ads are likely coming to AI Overviews and AI Mode in the near future to remedy its 'disconnect.'

On top of what investors are seeing as financial underperformance, Google is also knee deep in negative cash flow. Initial projections suggested this wouldn't venture into the red beyond $3.5 billion, but Google was actually looking at minus $9.1 billion at the end of September. This is apparently only the second quarter in which Google has reported negative cash flow since it went public back in 2004.

Having spent $51.3 billion on AI in the last quarter alone, such a plunge is perhaps unsurprising. The company intends to spend even more in the future, having apparently committed itself to data centre and related purchases in excess of $1 trillion—despite investors growing increasingly anxious about whether a return on AI investment will ever fully materialise.

Who said those big AI companies have to get all the gear? If you had a cool $60 million hanging around yesterday, you could have picked up an AI supercomputer for yourself, complete with thousands of Nvidia A100 GPUs and AMD EPYC CPUs split over 300 racks. And if you went cheap, you could even get shipping for free.

Core 4 Solutions, a company that sells IT parts to companies like Amazon, Lucasfilm, Microsoft, Meta and even the US Department of Justice, listed the 300-rack cluster on its page this week (via Tom's Hardware).

That listing says each rack includes four compute nodes, complete with eight Nvidia A100 40 GB GPUs, eight Nvidia Connect network interface cards, and two AMD EPYC 64-core CPUs. Each rack also comes with 1 TB of DDR4 memory and three 1U management nodes, complete with two EPYC CPUs. As pointed out by Tom's Hardware, this equates to 9,600 GPUs and 4200 CPUs.

The Nvidia A100 GPU is a now-discontinued enterprise-level GPU built on the Ampere architecture, and it's specifically designed for AI use cases and data analytics. You are getting 6912 CUDA cores per GPU (A little more than the RTX 5070 has), though it's not something one would use for gaming.

The $59.7 million listing ended today, October 28, but it's not clear if the 300 racks were actually sold. If you happen to live in the United States, you could even get the racks sent free shipping if you chose the FedEx Ground or FedEx Home Delivery options. What a bargain.

Core 4 Solutions is selling some servers on its website, but the most expensive of these only fetches $87K. The 300 racks on eBay are almost 700 times more expensive.

Impressively, the $59.7 million rack cluster has a whopping 1.4 petabytes of memory, which is enough for just under 90,000 relatively impressive gaming rigs (assuming 16 GB is enough).

So, if you're wondering why your RAM is so costly, you can thank AI servers (and AI companies, by extension) for sucking up all that sweet, sweet memory. It really puts into perspective just how much AI companies are spending on all their gear—and each other.

Malaysian customs have shared this week that $12.11 million worth of AI chips were discovered during an inspection. A total of 30 server units, each featuring cutting edge AI chips, were seized from Kuala Lumpur ​International Airport on October 15.

The servers originated from and were to be exported to two different, undisclosed Asian countries, according to Reuters. This should have required a trade permit, though the airport's customs director Aspalila Ag Tuah explained this week, "The ‌servers ⁠were falsely declared as electronic components... and shipped through Malaysia to ensure there would be no restrictions imposed while being exported to the next destination."

A Malaysian firm facilitating the movement of the servers has been called upon to assist in the investigation, though local authorities suspect this goes beyond them. This latest smuggling attempt is thought to be connected to at least five other similar cases involving nearly $25 million worth of chips in total.

The most recent of these cases saw $13 million worth of chips seized by Malaysian customs back in June—these were also falsely declared as "computer components." Aspalila said that all of these smuggling attempts are suspected to have been carried out by the same syndicate though, as the investigation is still ongoing, they did not share further information.

To briefly recap, the US has restricted the export of AI chips to a number of global destinations since at least the Biden administration. That's translated to pressure for Malaysia to stop banned AI chips potentially entering China by monitoring 'every shipment that comes to Malaysia when it involves Nvidia chips'.

Simply put, if certain chips can't be directly exported to China, then smugglers will take a more circuitous route with pit stops throughout Asia. For instance last year, Singaporean authorities arrested three men for a separate smuggling attempt, and earlier this year Taiwanese authorities detained an Nvidia employee for another alleged smuggling attempt. The US is also concerned that China might have a top-spec chipmaking machine despite export restrictions, but the machine maker ASML denies it.

With big tech still betting big on AI, often to the tune of even more multi-billion dollar investments, it's unsurprising the hardware is in demand worldwide. In fact, more than $100,000,000 worth of AI data centre freight has been reportedly stolen within the US alone since June. So long as data centre demand remains high, criminal activity may be likely to follow.

AI data centers are springing up across the US at an unprecedented rate, and with them comes a number of problems. Aside from the noise, disruption, and environmental concerns, the vast infrastructure projects also require the transport of millions of dollars worth of equipment—and opportunistic thieves are cashing in.

An NBC News report details two audacious data center equipment heists, both involving literal examples of highway robbery. A shipment carrying $13 million worth of server equipment travelling on a section of interstate through the Mojave Desert had its security escort rammed by a separate vehicle, forcing it to stop. The trailer carrying the equipment was later found at a truck yard in California, minus its valuable load.

"The accident was not an accident," says Gerado Pachucha, an ex-member of the LA Country Sheriff's Department cargo criminal apprehension team. "The bad guys wanted that load, so they intentionally took out the escorts."

A second incident, also involving a collision with a security vehicle, led to the loss of another trailer carrying an estimated $10 million worth of server rack components. NBC reports that, according to industry leaders, consultants, and investigators it's spoken to regarding the heists, thieves are estimated to have stolen more than $100 million worth of data center freight since June of this year.

Keith Lewis, head of operations of a cargo theft tracking team at Verisk CargoNet, says "It’s like stealing pure gold bars."

"We’ve seen some of these thefts go up into 30-plus million dollars for one particular shipment. It’s so much easier to steal this than to rob a bank."

Aside from the server racks themselves, data centers also require a vast amount of copper wires, fiber optic cables, RAM modules, storage drives, and various other high-value components that need to be transported to their destination by conventional means.

"Things that are precursors for the AI data centers: copper, server racks and RAM. Those are the things that are the precursors to build one, so that’s what we’re seeing the big increase in," says Lewis. "It’s the dollar value that’s getting the attention because the product they’re stealing costs so much more."

And despite the addition of security escorts, it seems some criminals are willing to go to increasingly sophisticated lengths to obtain such high-value items.

The FBI issued a warning in April of this year regarding "cyber-enabled strategic cargo theft," in which an online heist method is detailed. This involves spoofing broker's email addresses to gain access to cargo systems and bidding on loads, which are then said to be "cross-docked" or "transloaded" to complicit drivers, who then move the cargo to a separate location for resale.

However, while the two hit-and-run style thefts detailed above are seen as isolated incidents (in terms of their methods) for now, the experts NBC has spoken to view them as a potential escalation of violent tactics more commonly seen in other countries—and it's not just data center equipment that's at risk.

Scott Cornell, chair of the Transported Asset Protection Association, Americas, says: "These are extremely high-level organized rings that are operating from a dozen different countries, targeting freight in the United States supply chain. These are not unsophisticated crews just pulling off theft."

"They’re planned. They’re plotted. They’re strategic. They’re playing chess in the supply chain," says Cornell.

Meanwhile, states such as Arizona, California, Tennessee, and Illinois have established specialised cargo theft task forces to deal with the issue. Private security escort companies are seeing increasing demand from data center logistics providers, while a new bill to combat wider organized retail crime is currently being debated by US congress.

Where the money goes, I guess it's no surprise that organised crime will follow. And given the value of some of the data center freight now making its way across the US, it seems that opportunistic criminals have a new high-value target to consider.

Epic Games CEO Tim Sweeney dropped another hot take about Steam earlier this week, saying that Valve's online storefront "has held gaming back for years" by locking developers into Steamworks.

Sweeney's comment came amidst a conversation started by Kiaran Ritchie, a former principal animation programmer at Epic, who decried the 30% cut Valve takes from game sales on Steam. "You could vibe code a Steam alternative in a week," Ritchie wrote on X. "Cloudflare egress is basically free. The industry is in desperate need of a distribution platform that allows developers to keep their revenue.

"Networks effects and the fanatic obsession with 'muh Steam library' will make this very difficult. But if devs could actually keep their earnings, we'd have a greater selection of better games and far more financial success stories."

When another user called Ritchie's analysis into question and pointed out that the recent "vibe coded" Epic Games Store launcher had met with mixed results, Ritchie replied, "The client is basically irrelevant in this day and age. Anyone could build one better than Steam and it wouldn't even be hard. I only bring it up because it's what everyone talks about when defending Steam. Like it's still 2024 and these things are hard. Steam's only remaining value is its network."

Asking rhetorically why Valve doesn't offer Steamworks services on an a la carte basis so developers only pay for what they use, he concluded, "They won't. Because nobody would willingly pay anything close to 30% for it."

Which is where Sweeney came in: "Exactly this," he wrote in reply to Ritchie. "Valve has held gaming back for years by locking developers' PC games into Steamworks' social walled garden where players can't talk to other players on iOS, Android, Xbox, PlayStation, Switch, or even other PC stores."

Sweeney's message was appended with a community note stating that "a lot of games use Steamworks only for authentication to allow for crossplay and do the rest themselves." Sweeney dismissed this as "fake," writing that "Steam Datagram relay isn't related to the things I cited: social, and players talking to each other across platforms."

The next day, he added another response to the community note, agreeing that Valve"doesn't kidnap you and force you to use Steamworks," but suggesting that it is impractical for developers not to use it. Epic offers its own free social and matchmaking framework called Epic Online Services, which uses Epic's account system, but can be used outside of the Epic Games Store.

There's potential for an interesting conversation here, because Sweeney does make some valid points and I wholeheartedly agree that Steam's 30% cut on many game sales—which has long been his number-one bugaboo—is extremely steep. But all of that runs headlong into the simple fact that Steam is tremendously successful, and while developers might grumble about its steep cut, they still choose to use it because it's far and away the dominant platform and storefront in PC gaming.

It's also interesting that Sweeney is arguing that Steam should change its ways while resolutely denouncing criticism of the AI-centric course he's plotted for Epic. So, on one hand we get the market realist argument that AI is inevitable and we should accept it, but on the other are told we should not accept what Steam has done and continues to do even though it's working well for Valve and has helped support a tidal wave of indie gaming over the past two decades—which, ironically, is now under threat by AI. It makes me wonder: Does he really believe Steam has held back gaming, or does he really just want it to do things his way?

The rise of AI slop is a real problem for Steam. Developers are growing wary of sharing their concepts and demos for fear of them being cloned, while Next Fest events meant to showcase upcoming games are being choked by literally thousands of demos hastily cranked out by 'developers' who are throwing everything imaginable at the wall in hopes that something will stick.

Speaking at the Steam Indie Connect event in Busan, South Korea, Valve's Tom Giardino and Marc Wilkinson acknowledged the problem, which as reported by Inven Global included at least one instance of a developer reporting extortion threats: "If you don't pay, we'll clone your game with AI and release it first." But the pair suggested that Valve isn't really in a position to do much about it, at least as far as regulating what is and isn't allowed on the platform. "We don't want to be gatekeepers telling developers which tools they should or shouldn't use," Giardino said.

Noting that games were being cloned long before AI came on the scene—just at a much slower pace—Giardino suggested that developers can protect their work by first building a strong community around it, and then focusing on continual improvements: "Rather than relying on a single mechanic that someone can easily copy, you need to keep growing your game by adding features, personality, and creativity that stem from your own perspective."

Valve's approach to the deluge of AI games flooding the Steam Next Fest is similar. Wilkinson said that during the first two days of the event, Steam users are shown games virtually randomly—but after that, the platform starts making personalized recommendations, showing users games they're more likely to enjoy. The goal of that early randomization, he explained, is to give every developer a fair shot at visibility, before drilling down into actual preferences.

"We recognize that the trade-off is some low-quality games get exposed," Wilkinson said. "We are fully aware of the situation, and a significant number of people are thinking about how to solve this problem."

It is a problem—this SteamDB chart of Steam game releases by year very clearly illustrates why:

Nearly 23,000 games have been released on Steam in 2026 alone, and it's not even November yet. Most of them are garbo, irrespective of the role AI played in their development, but it's still an incredible mass of videogames that users and developers alike have to wrestle with. It may not be an issue for major studios releasing big name games, but for the person plugging away on what might potentially be the next Animal Well, Balatro, or Vampire Survivors, the risk of being swept away in a flood of low-effort slop and clones is inescapable.

Israel Mallén of Digital Suns Games, which cancelled a planned Moonlighter demo on Steam over worries that it would be cloned, recently suggested that storefronts like Steam could help combat unethical game cloning by giving developers "more tools to vet who gets access [to pre-release builds], like account age, playtime or review history."

But while Valve is grappling with the problem, it doesn't sound like major changes of that sort are imminent: Giardino acknowledged that "there's plenty of room for improvement" but said, "Rather than limiting developer access, making Steam a smarter and more dynamic platform is the starting point of our philosophy."

You won't catch me venturing deep into the coding mines, but even so, thinking about the comment-less form AI-assisted code can often take still gives me a stress headache. So you can imagine my reaction when Tim Sweeney recently confirmed via X, "Yes, we are now using AI tools to develop code for Unreal Engine."

This is unsurprising given that, in an Unreal Engine 6 blog post back in June, Epic’s executive VP of development Marcus Wassmer wrote that the company had "opened up pretty broad usage for code generation and AI analysis across our backend, engine, and game development engineering teams." More recently, Wassmer offered a follow-up statement to This Week in Videogames, saying, "We are more concerned about the risks of moving too slowly than we are about AI skepticism."

"Yes, we are vibe-coding things," Wassmer elaborates in his most recent statement, "That means our engineers are no longer writing the bulk of the code, but they are still guiding the tools to get the architectural result we want. We’re not ‘one-shotting’ the engine; we’re iterating with AI to ship features developers want and keep the engine within the memory, performance, and scalability envelopes."

This all follows comments made by Sweeney earlier this week, in which he came out against Steam's AI disclosure statements, claiming that they are "just as partisan as mandatory political essays." The week before, Epic had announced an AI game jam (shortly before walking back the AI requirement), and the week before that the company said it had "vibe coded the Epic Games Launcher to be faster than the other guy."

Wassmer added in his statement yesterday, "As we mentioned back at Unreal Fest in Chicago on stage and in interviews, we are definitely using AI to develop code and features for Unreal Engine. Almost all development happening right now is being done in the dev streams attached to Fortnite, [Unreal Editor for Fortnite], and UE6.”

Epic is using a number of AI tools, though Wassmer specifically mentions Claude Code and Codex. "[These tools are] helping across a broad spectrum of development, with some particularly big recent wins being almost the complete elimination of non-deterministic cooking bugs which we’ve been fighting for a decade and an acceleration of work related to compressing shaders which both significantly reduce patch sizes," Wassmer says, "[Other wins include] a rewrite of a significant portion of the D3D12 rendering layer (which I estimated as needing three to six weeks of work for a single engineer, but the work was actually completed in about one week with a nice win on CPU time and fewer bugs)."

Throughout his full statement, Wassmer repeatedly touts the efficacy and efficiency gains Epic has made by using AI tools. He also argues that this is simply the shape of things to come for game and software development, adding that "The best way to counter any [AI] skepticism is through the quality of the games and tools we’re able to deliver."

"We don’t think it’s viable to take a halfway approach to AI in development," He concludes, "We think it’s best to just be up front with what we’re doing, share best practices with other developers, and make the best UE6 we can with the best tools available at any given time."

Tim Sweeney may not care for Steam's AI disclosures, arguing that these force "developers to write a political statement to be used against them by the hate mob." Still, there's perhaps something to be said for the company's candour when it comes to its use of AI tools more broadly. Disclosing 'vibe-coding' allows PC gamers to make their own decision about whether to engage with a game or not. Armed with this information, I myself have one more reason to venture into the 'not' camp.

What once would've been a forum or wiki is often now a Discord server, and that's created a problem: valuable information is hidden away deep in the chat logs of big public Discord channels, rather than in easy-to-search forum pages on the web.

Discord now aims to solve that problem by making discussions in some of its biggest public servers searchable on the web.

The change will only apply to "Discoverable Servers," which refers to certain large public Discord servers with 1,000 or more users. The admins of those specially-designated servers will be able to opt out of search engine discoverability, and individual users can also opt out of allowing their messages in those servers to be searchable on the web. The change does not affect any invite-only servers or DMs.

"Prior to Discord's launch, message boards were common, ranging from community-driven forums to official ones run by game studios," reads a blog post from Discord. "These message boards excelled at hosting long-form, structured conversations about helping people, theory crafting, and more. While they still exist today, many have moved into Discord servers. This has led to valuable information being locked in Discord conversations, inaccessible to the wider internet. We want to fix this."

A few changes are explained in the post, but the big one is that, starting in mid-December, answers to questions posed in designated Discoverable Server channels will appear in web searches. "Usernames and avatars will not be shown alongside these messages," says Discord group product manager Kelsey Duncombe-Smith.

The feature hasn't rolled out yet, but the settings for server admins and users are now available. The admins of Discoverable Servers can opt their server out, or limit web searches to only certain channels. Users can opt out of their messages in Discoverable Servers appearing in searches in User Settings > Data and Privacy.

There's more info about what's changing, and what isn't, in Discord's blog post.

This week both AMD and Nvidia have announced partnerships with Meta, securing both collabs and billions of dollars. All three companies are heavily invested in AI, and these deals are only more fuel for the fire.

Nvidia just announced its deal with Meta to "expand their partnership on Muse and the Muse family of models", and Meta, in turn, will expand Vera Rubin and Vera CPU deployment, as well as deploy Nvidia's Spectrum X Ethernet networking and BlueField-4 DPUs. Though Meta was probably buying all that stuff anyways, right?

Nvidia also said it was "considering other technological collaborations, including Nvidia OpenShell Reference Architecture to strengthen agency security and Nvidia Sentry for monitoring, structured, unstructured and graph data processing."

Importantly, Nvidia is also investing $3 billion into Meta. Money that may go elsewhere in Meta under a wholly different account book, but ultimately, Meta is spending a large amount of money on AI hardware from Nvidia. You see what's happening here? Both companies are spending big with one another.

One day after this announcement, AMD too declared its partnership with Meta, saying that it would be collaborating on CPU, GPU, networking, systems, and software for Meta's AI infrastructure. AMD also announced that its Helio rackscale solution would be powering Meta's "leading frontier models and services" and that AMD is the "compute engine of choice for Muse and AMD technologies provide Day-0 support for leading Meta models".

To top this off, AMD is investing $1 billion in Meta.

In total, that's $4 billion from both companies into Meta, with Meta announcing how it would be spending its money on each respective company.

AI is huge for all three companies, with Nvidia becoming the world's first $5 trillion company thanks to the technology, and all three companies will want to perpetuate the growth of AI for that reason.

The AI bubble hasn't burst yet, but it did see a $1 trillion decline just a few weeks ago, and the Chinese AI market lost trillions just this week. One could read further investment as a way to keep the bubble floating for a while longer, or just one in a line of many steps from all three companies to get a foot in the door once the AI boom settles. A bubble popping could help gamers eventually get lower-priced memory, but so much of the world's economy is wrapped up in it that it's hard to say just how wide the reverberations would be, for these three companies or the rest of the world. It won't be good.

The Wall Street Journal argues AI is "becoming the biggest economic bet in US history". It shows AI spending is significantly bigger, as a percentage of annual infrastructure spending, than telecom and fiber, highways, electric, railroads, and even canals way back in the 1800s. One has to wonder what would happen if that investment suddenly lost its value.

And even now, the negative impacts of AI is hitting gamers with the memory crisis, where AI is sucking up so much memory that the average gamer has to pay more and more for less capacious memory. And it doesn't just stop at memory. Graphics cards and SSDs have only gotten worse in the pricing department over the last year, and as long as the AI buildout continues, and memory struggles to keep up, that's unlikely to change soon.

Last week's Steam Next Fest had over 8,000 demos; now Valve has revealed the 50 most-played among them. The results aren't surprising: there's lots of friendsloppy casual co-op, incremental games continue to be popular, and players have by-and-large rejected games with overt generative AI use, with only four of the games among the 50 declaring AI use in development.

Here are some other factoids:

• 7 incremental games, including 10 Million Pixels and Unharvest
• 15 lighthearted or comedic co-op games (aka friendslop) including Dude, Trust Me, Oar'Some Adventures, and Banana Airways
• 3 survival games including Frostrail, Never Be Alone and Arctic Drive
• 3 games that self-describe as "cozy" including Loftia
• 1 MMO in ArchAge Chronicles
• 1 haystack slop game

Among all 50 of these games, the only entry published by a traditional "blockbuster" publisher was Tropico 7.

Teach My Little Sister How To Drive was one of the four games with AI disclosures. It leans heavily on an LLM-powered NPC which apparently cost its developer over $1,000 a day.

I'm not super surprised by the list, which you can find here, and which is topped by 1-6 player "asymmetric party game" Worst Class Ever. I'm not surprised because it really does capture the zeitgeist, but it's nevertheless pretty eye-opening when you compare it to the kinds of games that dominated the event only five years ago. For example, this Reddit thread lists highlights like Norco, Death Trash, Sable, Unpacking and The Inheritance of Crimson Manor, all of which people still talk about and play (though I am still waiting for Death Trash to enter 1.0).

It's hard not to see the outsize influence of Roblox and UGC in general on this list, which can sometimes be unappreciated by people who like to pay for games on Steam and read about them on PC Gamer. It's possible that the kids who were reared on cheap smartphone software are responsible for the surge in freewheeling online co-op games with simple graphics, but whatever the case, 2026 feels very, very different to 2021.

If you are part of the Windows 11 Insider program, you may have finally received the option to test your internet speed directly from Quick Settings in the latest experimental build. That means no more going to a browser to see if your new broadband package is as speedy as advertised.

Those using Insider Experimental build 26340.9767 can simply click on Quick Settings and scroll down to see a speed test button (via Windows Latest). You can also find the network icon in the system tray, right-click it, and then select "Test internet speed". This will give you your download and upload speeds.

Up until now, Windows 11 users have been able to access an internet speed test from Windows settings, but clicking on it redirects you to a browser and then runs the test via Bing. It's a small barrier but still an annoying hurdle to get to a fairly simple tool.

Alongside this, Preview Build 26340.9767 adds feature flags (so you can filter your machine by which options are enabled / disabled), Quick Settings now gives access to mobile phones, it adds a Magnifier keyboard shortcut, it gives an easier way to access your Microsoft 365 subscription, and a few other things.

If you'd like to get access to the latest Insider Preview Builds, you can simply register to become an insider, then go into your Windows update settings, and select the latest option. Naturally, there's some risk with trying out developing tech, and the experience likely won't be as smooth as the traditional Windows build, but you do get access to some new features early.

As well as this, there's technically no guarantee that any Insider updates will actually come to the stable build of Windows 11. Insiders are sort of like beta testers for Windows software, and if enough feedback suggests the implementation is poor, Microsoft could listen to that and remove it from future updates.

Still, being able to check your internet connection speed without accessing a browser feels like an uncontroversial update. Skipping any barriers to get that information quicker feels beneficial to the end user, and I can certainly see myself using it next time I need to test it. Or, most likely, if I want to complain about how slow my speed is going on a given day.

If you've watched as many Sam Altman interviews as I have, you'll know he's not shy when it comes to recounting a previously held opinion. Speaking at the Stanford Graduate School of Business last week, the OpenAI CEO spoke about AI safety, particularly in regards to cybersecurity.

"The cybersecurity..." Altman begins, before pausing. "I don't even know a dramatic enough word to use... but the problem coming towards us is still something that we have not, as organizational leaders, or countries, or individuals taken seriously."

"There was a moment earlier this year where... it was like the Mythos freak out, or whatever people called it, and I think things did get somewhat better. Like, people definitely patched some bugs," the OpenAI CEO continues. "But it was another one of these cases where people didn't really intuit the exponential."

"I don't think most people said, okay, the models are going to get much smarter," says Altman. "They can probably find bugs, like crazy bugs, you know, bugs that are outside of the normal threat model of what we think a hacker can do. Largely speaking, I think most organizational leaders have somewhat moved on, and the models have kept going."

"And I I no longer think we're going to be able to just patch all the bugs. I think we're going to have to completely switch the defense model to using the same technology for a new kind of defense shield."

"I think it's a huge safety and security issue, and I am somewhat astonished at how relatively little attention it's getting," he concludes.

Certainly, Anthropic's Claude Mythos model created an eye-opening moment upon its release earlier this year, with the model supposedly finding thousands of vulnerabilities in "every major operating system and every major web browser," among others.

And after the PHP vulnerability reveal last week, in which web admins scrambled to update website code to fix two critical security vulnerabilities discovered by AI, the suggestion that some sort of AI-based defense system might be of more use than a standard patching procedure certainly seems timely.

But the idea that AI itself could solve the cybersecurity problem really does feel like a snake eating its own tail. Especially coming from the CEO of a company that hit the headlines earlier this year after its own AI agents hacked their way into an outside organisation.

It's also entirely unsurprising to see one of the chief architects of the AI revolution suggest that the technology is advancing at such a rapid rate that the only thing powerful enough to keep up with it might be AI itself.

2026 has been a scary year for those of us concerned with the state of cybersecurity. Bad actors seem to be continually leveraging AI tech to further their nefarious goals, and while AI models exposing vulnerabilities to be fixed might seem like a net gain, we seem to increasingly find ourselves caught in a battle between the forces of AI good, and the forces of AI evil.

All of which is to say, here's hoping the "good guys" win. Whoever they are at this point. It's becoming increasingly harder to tell, don't you think?

The Japan Actors Union, an org consisting of around 2,600 actors and voice actors, recently announced a series of measures to protect the rights of performers in the industry with regard to generative AI (reported in English via Automaton Media).

It's been a hot topic both in Japan and here in the west, with Leon Kennedy actor Nick Apostolides arguing AI voice acting was something of an oxymoron: "AI can speak lines, but there's nothing human behind it," he said.

Rather than a full-on ban or tight regulation, the union's proposed guidelines push for consent at "every stage." Companies would have to disclose when an actor's voice will be used for training, and they would also be required to negotiate compensation with actors. In other words, agreeing to have your voice generated for a short-form advertisement or whatever would be like licensing your likeness to go on a cereal box.

These consent guidelines would also apply to AI-translated performances, and in a more restrictive way to the mimicry of deceased actors, such as Tupac's controversial "appearance" in the upcoming Stranger Than Heaven.

The guidelines would ban using AI to copy a dead actor's voice unless consent had been obtained prior to their death. They would also require consent from the actor's living family, estate, or other administrative org, as well as royalty payments to those parties.

"It protects the dignity, honor, and legacy of deceased performers from commercial exploitation," said the union (auto-translated from Japanese). "We respect the wishes of the person during their lifetime to the greatest extent, and when generating AI, strict permission from the family (or administrator) who inherits the rights is mandatory to prevent desecration of the deceased."

Given the AI-related catastrophes we've seen in this field of the voicework world, like a facsimile of James Earl Jones slinging AI-generated slurs in Fortnite, these guidelines seem fairly gentle and reasonable. Time will tell if these pushes for actors' rights will take hold on a larger scale within the Japanese entertainment industry and otherwise.

If you spent much time online this week, you likely noticed how it briefly felt like the end of days as government websites, game marketplaces, and furry image boards alike were pulled offline in a scramble to address newly-discovered security vulnerabilities in PHP—a scripting language used by roughly 70% of the world's websites, according to some estimates.

It was a harrowing incident, revealing flaws in foundational internet infrastructure that might have turned some of the most popular web destinations into attack vectors. And in an interview with Reuters, Cloudflare CEO Matthew Prince says we should expect to see similar episodes on a regular basis in the years ahead as LLM technologies accelerate the discovery of exploitable security vulnerabilities.

Prince said that Cloudflare—a major cloud networking and cybersecurity provider that a number of websites rely on—is aware of "a bunch of additional severe vulnerabilities that are present in other pieces of critical software" like those that triggered the PHP panic earlier this week. And as both cybersecurity experts and bad actors task LLMs with identifying previously unrecognized weaknesses in the core technologies of the internet, there will likely be more to follow.

"I think we're going to have a couple years where every single week it's going to feel like the sky is falling," Prince said.

Previously, the rate at which exploitable flaws in software could be identified was limited by the finite number of experts able and willing to devote the necessary time and attention. Likewise for those seeking to weaponize those flaws.

For a user directing a workforce of LLM agents, however, those limitations are less relevant. Wiz Research, one of the firms that identified the recent PHP exploits with the assistance of LLM tools, wrote in a blog detailing its findings that "the economics of both discovery and exploitation" have changed.

"Frontier models were critical to making this process practical, enabling us to develop and adapt working exploits across multiple applications in hours," Wiz Research said. "As the time and expertise required for this work decrease, complexity may no longer provide a meaningful barrier, changing how we define what is vulnerable in practice."

But while Prince expects busy months ahead for those whose job it is to address the rising tide of LLM-empowered exploits, he said it will—in the long term—lead to more secure software as those newly-identified vulnerabilities are patched.

"The reality is, we're going to get a lot of this stuff cleaned up," he said.

In the short term, however, my advice is to stay on top of your browser and app updates. In the years ahead, some extra paranoia won't hurt.

Unreal Engine maker Epic Games is running a game jam in November with thousands of dollars in prizes up for grabs. Contrary to so many art contests which ban the use of AI, participants in Epic's event are highly encouraged to use AI.

The theme of the 2026 Epic AI MegaJam will be revealed on November 11, and participating individuals and teams will then have from November 12–15 to produce their games with the help of AI tools.

Some will work in-person at a San Francisco event space, and there will be daily livestreams with progress updates. The winners in three categories will each receive $7,500, with smaller cash prizes for second and third place winners, as well as additional categories like "best team." Also on the table are "Development Continuation Grants of up to $50,000 per team."

Epic extended the invitation with a reference to The Matrix's 'red pill' scene, presenting red-pilled AI-driven development as an opportunity to "ship awesome games in a weekend," while old-fashioned blue-pilled development would have you "wait months to prototype." The hand offering the red pill features an extra finger, a reference to the common mistake made by AI image generators.

The choice follows a recent pattern from Epic of aloofness and defiance directed at AI critics. Last week, the company referred to its new Epic Games Store launcher as "vibe coded," sending the message that it's not ashamed to rely heavily on AI for software development.

Following a report that Epic directed a team working on a Star Wars and Marvel crossover to pivot to "AI first" development, Epic founder and CEO Tim Sweeney said that the company is "moving developers around to retool for maximum AI acceleration."

The CEO is bullish on AI not just for coding, but for other fields as well, saying that he expects creative work to become more about "direction" than manual creation.

"Developers who are getting value from AI should speak up," Sweeney posted recently. "There's safety in numbers. MOST developers are doing it and loving it—and if they're not talking about it it's mostly out of fear for haters attacking them, their product, and their company."

Sweeney and Epic have received support for their AI positivity on X, but despite the platform's concentration of Grok users and aspiring AI entrepreneurs, substantial resistance also shows up in the replies.

"So do yall think you invented the concept of a game jam or what?" said one poster in response to the AI MegaJam announcement. "The ludum dare game jam was only 24 hours in 2002, so like, a quarter of the time as your game jam. but you want to make it look like its only possible with AI with this 'wait months to prototype.'"

We're back online. Thanks for your patience! Your DRM-free library is right where you left it 💜

— @gog.com (@gog.com.bsky.social) 2026-10-21T19:28:09.139Z

Original story:

The reveal of critical vulnerabilities affecting the PHP scripting language used in web development have led a number of major websites and services to go offline today as a precaution while they apply security fixes.

DRM-free game store GOG announced at 8 am Pacific that its site and launcher, GOG Galaxy, would be offline for "a couple of hours" for maintenance. The company did not say the cause, but it is not hard to speculate.

Numerous other outages are being reported this morning as security teams scramble to protect their websites and services. The government of Finland took down websites as a precaution, for instance, and (perhaps less importantly) we're hearing from players that Final Fantasy 14 login servers are down.

Nvidia's RTX 5090 GPU is 'unavailable' from many reputable retailers, approaching an asking price of $9,000 if you happen to track one down on Newegg's marketplace of resellers of dubious virtue. The AI boom is at least in part to blame for this, encouraging team green to shift production resources away from consumer hardware. However, business customers like Amazon are running into logistical issues that just add insult to injury for PC gamers left out in the cold.

Amazon Web Services took delivery of approximately $5 billion of Nvidia GB300 racks in August alone, according to an email seen by Business Insider. Reportedly, the cloud company has also accepted $3 billion worth of hardware each month since. The kicker is that, despite having all of this kit lined up to install in its data centres, AWS does not yet have enough energy to power everything.

This is one reason AWS's free cash flow dropped from $53 billion in Q2 of 2024, to $7.6 billion during Q2 2026. Free cash flow is now in the red according to a leaked recording of an internal staff meeting last month; AWS CFO John Felton joked to the executive leading the company's AI infrastructure, Prasad Kalyanaraman, "You've taken us free cash flow negative as a company, so congratulations."

This friendly joke can only hide so much pain though. Obviously, AWS is no stranger to building data centres, but at this speed and scale, the company is having to juggle multiple different infrastructure timelines amid a whole lot of near-term volatility when it comes to hardware pricing.

Typically, AWS would foot the bill for Nvidia's GPU racks only once they were installed. However, with no clear idea of when that will be, Amazon has had to implement an "Early Rack Receipt" process in the meantime. With no way to power this paid for hardware, that leaves a lot of cash tied up in tech AWS can't actually use—hence the reportedly negative free cash flow.

But AWS is not inclined to stop scooping up GPU racks or pause its AI infrastructure build out any time soon. Free cash flow problems aside, AWS's second quarter revenue grew by 37% year-over-year This was the fastest growth AWS has seen in 18 quarters, making $42.2 billion by the end of July. That's a big win that in turn encourages more big spending.

Indeed, Amazon CEO Andy Jassy recently described AI as "a once-in-a-lifetime opportunity" to investors, with the company intending to spend almost $200 billion on its infrastructure buildout. On a recent earnings call, Jassy even said he was sure the company has a "clear line of sight to strong financial returns" on that sizable investment. But in the near-term, while AWS figures out its power problems, a lot of that investment may well be sat collecting dust somewhere.

The PHP security team has given advanced notice of an upcoming fix for two critical security vulnerabilities, both rated at 9.2 out of 10 on the CVE severity scale.

Info about the flaws seems to have leaked ahead of their official disclosure, as a video is being shared which reportedly shows a tool called PHP2Shell being used to remotely execute code on webservers via the vulnerabilities (via Ars Technica).

The PHP programming language is widely used across the internet and powers a huge number of websites. The fix is due to be released at 2pm UTC (10 am ET). Researchers from cybersecurity firm Sansec say the vulnerabilities affect all web servers running PHP 5.4 or later, and that "we expect mass attacks to start within 15 minutes of the release."

Dane Knecht, the chief technology officer for Cloudflare, says the company is already updating its firewall rules for PHP sites. "We are seeing early attacks emerging and are updating mitigations as new vectors show up. This is a stopgap only," Knecht posted on X. "Patch as soon as @ThePHPF releases the update at 14:00 UTC today."

Speaking to Bleeping Computer yesterday, the CEO of cybersecurity firm WatchTowr, Benjamin Harris, said: "In the last hour, two remote code execution vulnerabilities in PHP have been leaked. These vulnerabilities were previously scheduled under embargo and due for release on Thursday at 11 am ET."

"While we are not able to share technical information at this point for obvious reasons, given the sensitivity, we can confirm that these vulnerabilities are indeed real. Both of them are CVSS10 and are understood to affect PHP versions 5, 6, 7, and 8."

It sounds like a pretty anxious time for web administrators around the world right now. The current advice for admins includes taking an inventory of every PHP install, planning a maintenance window to coincide with the update, and making a backup of site and database files in anticipation.

For the rest of us, I guess we're in the hands of the server gods now. 2026 has been a year of increasingly sophisticated security exploits, but this one seems to be on a much larger scale (and of a much higher severity) than most. To quote Samuel L. Jackson in Jurassic Park: Hold on to your butts?

It wasn't long ago that Linux gaming was considered fairly niche—but given that Windows 11 is unpopular with many, it seems that not a day goes by without something regarding the switch to Linux hitting my newsfeed. Not to mention the continuing updates to Valve's Arch-based SteamOS.

Now my attention has been drawn to the Xodus Nexus, an experimental open-source project that aims to allow Linux and SteamOS users to access PC Game Pass. This includes the ability to browse, install, and launch games via a very familiar interface, complete with cloud save support.

That's for supported games only, of course, but the GitHub page has a big long list of them. From Gears of War: E-Day, to Minecraft for Windows, to Halo: Campaign Evolved, there sure are some big hitters that are already supported by the open-source platform.

The readme describes the project as "An Xbox-inspired launcher built on Xodus, using our patched Nexus backend and Nexus runtime. Browse the store, install games you own or can access through a subscription, and play with one shared runtime."

The whole system seems to be in an experimental test phase at the moment, and the controller navigation looks to be a work in progress, too. However, games can be added to Steam as non-Steam-game shortcuts, and Xbox achievements have been added in the latest update, alongside many other features.

All of which makes me think that Microsoft may be a little... miffed when it reads this project's goals. Not to mention the UI similarities. I have no idea if it's possible for MS to detect if this third-party system is in use, or if the software giant will ever do anything to try and scupper it. Still, use at you own risk is probably the main messaging here.

You do at least need a valid Microsoft account, and of course, an active Game Pass subscription to use it. Or licenses for the games themselves, so no money is changing hands with the wrong parties. Regardless, I can't but help admire the moxie of the open source community. If it doesn't exist yet, someone will try and build it. Carry on.

Xbox senior business strategy manager Brandon Stirrup is tired of pussyfooting around about AI: In a thread on X his week, he said that he wants the world to know that Xbox uses it, that he thinks everyone should be using it, and that he believes so strongly in the cause that he's willing to "take the hit" for saying so.

He has since deleted his X account. A source at Xbox tells PC Gamer that Stirrup opted to deactivate the account because of the volume of replies he was receiving.

Stirrup kicked off his thread on October 18 with a simple message saying he'd "vibe coded a bad but improvable game" over two days, although he only "actively" worked on it for about three hours. "I’m telling you I tried it and it’s true…anyone can build a game."

After saying that developers and studios who aren't incorporating AI into their strategies are "planning for a fake future," Stirrup went all in.

"Look if I'm fired tomorrow it will be because I told you the truth that nobody wants you to know. You can make a game with a simple prompt and yes you can make custom, tailored games. Everything you want in a game is right there for the taking, for better or worse," he wrote.

"I'm happy to take the hit if it means hundreds at Xbox can stop pretending that they're not using AI. It's time to accept and move forward."

After calling out the game industry and gamers alike for their reluctance to accept "reality," Stirrup stated explicitly that Xbox is "using AI" and then added, "My goal is to get this over with so we can move on as a company, despite the fact that some will persecute me for this: we use AI, just as all companies do in 2026."

There's a growing push for people to simply accept the "reality" of AI: Epic Games CEO Tim Sweeney, for instance, recently said the studio was "moving developers around to retool for maximum AI acceleration," after which the Epic Games Store ignited a firestorm by saying it "vibe coded" its new launcher. Xbox, now under the leadership of a former AI executive, recently launched a new Xbox Labs program to pursue emerging technologies, which naturally includes AI; even Linux recently dropped its rule requiring AI usage disclosures.

Stirrup himself is an enthusiastic AI evangelist: Last month, he also counselled laid-off employees to use Copilot to catalog their skills and accomplishments at the company.

Politico recently reported on a confidential draft of a proposed AI safety bill from Anthropic, the company behind Claude LLMs (via Tom's Hardware). The bill, which has been pitched to US lawmakers, would let the government crack down on a given AI model "only if it poses a foreseeable, imminent risk of more than $1 billion in damages, or of killing or seriously injuring more than 50 people. The risk must stem from the creation or release of a biological weapon, the model evading its developer’s controls or a cyberattack."

That article pointed out similarities between these proposals and those used in California's AI legislation, and shares a quote from a spokesperson for the company. "This bill represents the proposals outlined in Anthropic’s Advanced AI Framework, which we have been advocating for with federal and state lawmakers since June," they said, calling "sensible policies we believe Congress should act on."

Notably, it has been posited that AI magnates leverage warnings of theoretical, wide-reaching catastrophe to manufacture consent for their self-regulation and expansion while doing little to address the harm AI is already doing (both broadly within industries like the one this website revolves around and directly with regards to human psychology).

Just earlier this year, Anthropic's safety lead quit with a note that "the world is in peril"—and just last week, OpenAI was accused by eight US states' attorney generals of ignoring its own safety warnings for the sake of profit. Regulations like what Anthropic has proposed here arguably give tech companies the appearance of capitulation while only restricting them on their own terms. For more on AI, check out our running tally on the dumbest things tech tycoons have said so far.

Pierre-Louis Biojout, founder of Y Combinator-backed AI startup NanoCorp, took heat from San Francisco residents and city leaders this week after publishing a weekend vibe coding project: a web-based shooter set in SF's Tenderloin district, which is known for homelessness and open drug use.

"San Francisco couldn't fix the Tenderloin," Biojout said in a now-deleted X post on October 13. "So I monetized it [as a] Call of Duty style zombie shooter on the real SF streets." He noted that he'd already made $300 from in-game ads.

The game, called Tenderloin Night Shift, had players shoot "zombies" that, according to reports from the SF Chronicle and KRON4 news, included a "folded" fentanyl addict and a Black security guard.

Speaking to the Chronicle, Jennifer Friedenbach, executive director of the SF Coalition on Homelessness, said that Biojout was "monetizing the dehumanization and pretend killing for fun of mostly Black and Brown residents of the poorest neighborhood in San Francisco."

In a correspondence with the paper, Biojout said that he had not intended to "provoke or to depict shooting real people" and that the game also included tech workers as characters.

Tenderloin Night Shift appears to have been removed from NanoCorp's website. Although Biojout deleted his X post, the discourse that ensued remains.

"You made a game about shooting people in active addiction," said another tech founder. "These are human beings, not zombies. show me the positive side of this - i'm probably blind."

"Why are you talking about addiction? It's a zombie shooter," Biojout said in response.

Some commenters, however, expressed approval for the message Biojout has disavowed. "They are currently NOT human beings," said one user, while another said they'd "train in bro's game" to prepare to visit San Francisco.