I may not have mentioned this yet (I'm pretty sure I mentioned it in one of my first PC Gamer articles) but I have a law degree. That has made me love the minutia of the legal system but also keenly aware of how complicated and drawn out it can all get. Though the United States' enforcement of anti-trust principles as of late is likely better for the consumer and broader market, the latest hearing isn't the end of the case.

United States v. Google LLC (1:20-cv-03010) began proceedings in October 2020 and was decided in August of this year. This is a different case from the confusingly titled United States v. Google LLC (1:23-cv-00108), which started in 2023, and had its closing arguments in November 2024.

The latter case is specifically about Google's control and alleged monopoly over its advertising, whereas the former is about its search engine. Through deals with companies like Apple to be the default search engine on the iPhone and Google's ownership of the popular mobile OS Android, Google has become the search engine giant it is today.

There's a reason why it's a popular joke to jokingly insult the 'five users of Bing' or how behind the market Microsoft's Edge is. That browser is known as the app you use to get Chrome, then never touch it again. In fact, Windows periodically sends reminders that 'you can just use Edge, you know?' and yet many don't. I write this typing on Google Docs, through Google Chrome, so maybe I'm part of the problem.

Fundamentally, most browsers aren't too dissimilar from each other—yet Chrome holds almost 70% of the market share on a consistent basis. This lawsuit took aim at the browser, and after years of deliberation, the DOJ has proposed that Google be forced to sell Chrome, alongside a few other efforts to stop Google from further monopolising the market.

Judge Amit Mehta, who presides over the case, has ruled that Google acted illegally to develop and maintain its monopoly. As well as advising that Google sell Chrome, the DOJ proposed it be barred from reentering agreements like those that made it the default web browser, whilst also recommending it share its data with rivals.

It's also been recommended that Google either sell off Android or avoid making Google the default option on Android devices. The DOJ makes these recommendations to Judge Mehta, where Judge Mehta can mediate and come to an agreement.

This all seems like great advice to stop a company that has acted in a way that the court deems to be monopolistic. Controlling such a huge part of the market not only discourages others from entering but can push those in the market out. Once a company has a big enough control over a resource, like all the digital real estate Chrome has, it can essentially make its own rules.

Outside of court orders like this, there aren't ways to hold huge corporate entities accountable when they are acting against the interests of their users. If Chrome, Edge, Opera, and more were equally viable choices, you could afford to swap from one to the other if you didn't like any decisions made.

Of course, selling Chrome won't shrink its user base overnight. Though the other actions recommended by the DOJ would help with that too, it's important to note that the litigation against Google is more than just a singular case, it's one of multiple antitrust suits levelled against the company in the last few years.

The worry isn't specifically and exclusively about Chrome, but all the actions made to keep it where it is today. Google as a company doesn't appear to be trusted to handle something like Chrome in legal ways.

Published on November 21, just a single day after the DOJ proposed Google sell Chrome, Kent Walker, the global affairs and chief legal officers of Google and Alphabet (Google's parent company), took to the Google Blog to declare that this decision "would hurt consumers".This piece makes the argument that the proposal would endanger the security of users by undermining the quality of the services and disclosing Google's research to "foreign and domestic companies".

It also argued that Google's investment into AI would be chilled by this decision, effectively warning that Google's own advancement of AI would be held back as a result.

The blog also argues that it would not only hurt casual consumers' access to Google search but would hurt companies reliant on Google, like Firefox. Finally, it argues this is a slippery slope and an overreach of the government on "your online experience".

None of these arguments, on their face, are necessarily wrong—but I'd argue they're lacking a little in nuance. Selling off such a huge entity could indeed leave consumers open to bad practices, both technically and ethically, from the potential buyer. It is also true that Google's research into the likes of Google Gemini would make less sense without owning the broader ecosystem. Also, with so many browsers being based on Chromium, Google's open-source web browsing project, there's a chance for security or quality problems as a result.

However, that's also part of what the current process in the lawsuit is for. Judge Mehta has argued Google acted illegally and Google does have the chance to appeal, which it has been suggested it will do. If it fails to file as such, a further trial will be held in April next year to come up with an answer to remedy the alleged wrongdoing.

In this case, Google's arguments will be heard and an answer will be found. Many of the arguments brought forward by Google could be remedied through further litigation and aren't themselves arguments against the suit. It's important to remember that the DOJ's recommendation isn't law, it's merely a suggestion based on the facts of the case, and other remedies can be found.

Though we have many arguments and a decision, this case is anything but final and it's hard to have too concrete an understanding until a later date. Though the breaking up of parts of Google seems to have been necessary for some time as argued by Judge Mehta, the specifics of how that will happen will dictate how successful the potential selling would be.

Chrome is a major part of much of how Google currently operates and measures will need to be taken to not lose many fundamental parts of the internet seemingly overnight. A heavy-handed approach was needed for much of the court case to even bring this suit against Google but a finer touch is needed for those specifics. Though we won't understand how fine that is for some time.

Microsoft has been logging many of the various known issues on its Windows release health page, and the most recent entry relates to Auto HDR. Some users have been experiencing incorrect game colours with "certain display configurations" and even full-on crashes in some games.

Microsoft has applied a compatibility hold on devices with Auto HDR enabled, meaning that these devices will not be offered the option to install 24H2 via the Windows Update release channel if they haven't installed it already.

However, that's far from the only 24H2 bug doing the rounds this holiday season. Second on the list is an entry mentioning audio issues with a "limited set of devices from one manufacturer" namely devices using Dirac Audio with cridspapo.dll.

Again, Microsoft says it's put a compatibility hold on affected PCs—but both myself and my colleague, Nick Evanson, have had audio device issues since installing the update, and neither of us uses Dirac Audio. In my case I've had audio devices disappear, refuse to switch over (which is fun for meetings!) and odd crackling.

All of this is directly post-update, which for me also broke many of my drivers and required me to reinstall them manually. Oh, and my Windows notifications keep disappearing and reappearing en masse, too.

Polling the hardware team for odd PC experiences since installing the update reveals yet more strange behaviour. Our Jacob Fox has experienced animation glitching when minimising and maximising individual windows, which a quick Google search reveals seems to be an ongoing issue for many users.

He's disabled animations entirely to fix it, so spare a thought for poor Mr Fox this holiday season with his barebones Windows experience.

Nick has ongoing issues with Chrome, which (directly after the installation of 24H2) occasionally freezes page rendering halfway through, leaving the bottom half stuck in place but the top half still active. And as for myself, it's not just audio and notification issues that have me scratching my head.

I noticed odd stuttering and hangs in games that ran smoothly pre-update, including Diablo 4 and No Mans Sky, both of which usually run consistently at over 100 fps on my machine. Spending an evening scouring forums for a fix (not exactly my favourite way to spend a night off), I noticed that some people were recommending turning off fast startup and disabling hibernation to solve the issue in certain games.

And wouldn't you know it, that's worked for me too. These are anecdotal experiences of course, but I'd say it was worth a try if you've experienced odd hitching and micro stutter issues since installing the update yourself.

All this, and I haven't even got to the Ubisoft debacle, in which 24H2 was prevented from being sent out to machines with certain Ubisoft games installed on them, like Star Wars Outlaws and Assassin's Creed: Valhalla, in order to prevent crashing. Ubisoft has been releasing hotfixes to solve the problem, but certain games are still yet to receive a fix.

Previous issues have included the 24H2 audio jump scare bug, the Western Digital SSD BSOD bug, and a borknado of other pre-public launch problems.

So, it seems 24H2 continues to cause headaches for many. I can't remember a time a major Windows update didn't cause widespread issues, but even so. this latest patch seems to cause more than most.

So here's hoping Microsoft squashes these bugs sooner rather than later. I have games to play over the holidays, you see, and I'd rather not spend my time off scouring the forums for yet more hackaround fixes.

Today, for day 10 of its "12 days of OpenAI" Christmas lead-up event, OpenAI has gone further than I expected by unleashing its chatbot from the tethers of the internet and into the telecom airwaves. Yes, you can now call ChatGPT or WhatsApp message it. Unfortunately (or fortunately?) you can only do the former if you live in the US, but if you live elsewhere you can still do the latter.

The number is 1-800-CHATGPT, and yes, it's toll-free. Which is how OpenAI is marketing it, by the way: "Call toll free, 1-800-ChatGPT." Nice. That's 1-800-2428478 for those unfamiliar with traditional keypad texting.

OpenAI's Keven Weil says that this is all about "making it [AI] as accessible as possible to as many people as we can."

As you can see from the image below, I found ChatGPT WhatsApp to work just fine. The QR code I've slapped on to the right-hand side is taken from OpenAI's video. If you scan it, it opens up the WhatsApp app and takes you to a conversation with ChatGPT.

I can't test the phone calls as I'm not based in the US, but judging from the phone calls the OpenAI reps make in the video, that seems to work just fine.

I'm slightly jealous of those of you living across the pond after watching that video, too. If I could get over the initial embarrassment of doing so, I reckon being able to speed-dial ChatGPT could come in quite useful. A bit like using Siri, but... well, more conversational and better (sorry, Apple). Plus, you could even do so on a dumbphone.

Just don't go using it as replacement for, you know, actual human contact. It's not a conscious human being. Maybe try hanging up on it a few times to get your mind used to that fact. It's not rude, it doesn't feel anything. Don't believe me? It even says so, unprompted, in our chat above.

So there you go, that's day 10 from OpenAI for you. There've been some smaller announcements from OpenAI over this 12-day bonanza, but some bigger ones, too. On day three, for instance, the AI company unleashed Sora, its video generation model. And now we have this: ChatGPT phone calls and ChatGPT on WhatsApp.

I wonder what the last couple of days will bring. I'm not sure much could top this, to be honest. Well, not much that doesn't have me breaking open the Skynet Defence Manual. And that wouldn't be very Christmassy, would it?

As reported by the Financial Times, the tech consultancy Omdia estimates that Nvidia's biggest purchaser of Hopper chips in 2024 was Microsoft, who bought 485,000 of them, this being over twice as many as any other company. Meta, for example, bought 224,000 of them, Amazon bought 196,000, and Google bought 'just' 169,000.

Surprisingly, though, two of Nvidia's biggest customers were Chinese ones, these being ByteDance (of TikTok fame) and Tencent (of numerous videogames fame). According to Omdia, these each ordered about 230,000 Hopper chips. These won't have been the most powerful ones Nvidia has at its disposal, though, given US-China export restrictions.

In other words, despite export controls, Chinese companies received more Nvidia chips than companies such as Meta, Amazon, and Google this year.

Apart from all the China biz, there are two other things that strike me about these numbers. First, and very simply: Holy moly does Nvidia churn out a lot of a AI chips. Second: Holy moly does Microsoft buy a lot of them.

Nvidia's part kind of goes without saying. The company is firmly cemented as the king of the ever-expanding AI infrastructure castle. So much so, in fact, that the company's CEO, Jen-Hsun Huang, has the chutzpah to claim that "just like we generate electricity, we're now going to be generating AI" in "AI factories".

As far as Microsoft's concerned, while I'm a little surprised to find the company more than doubling the purchases of any other, it also makes sense, especially given the partnership with OpenAI.

The AI industry can seem a little confusing when you start to look into how all these different companies relate to each other. But we shouldn't forget that while OpenAI technically isn't the biggest company in the AI space, most of the companies that are bigger actually rely on OpenAI's software and models and have partnerships with the company. So much is true for now, at least—although newer entrants such as Anthropic and Musk's xAI could make inroads.

And if we're talking OpenAI partnerships, Microsoft's is the one. To date, Microsoft has apparently invested $13 billion into OpenAI and is the exclusive provider of the company's cloud computing services. This partnership grants Microsoft all kinds of benefits, such as OpenAI model integration with Bing, Microsoft 365, Copilot, and so on, not to mention the ability to rent OpenAI-clad Azure servers out to customers for private or bespoke AI research or services.

Oh, and there's the simple matter of monetary ROI. But it's surely crude to speak of such things (profit motives in such a civil society? I think not).

People want AI and OpenAI is the biggest software-level solution, so Microsoft heavily invests in and partners with OpenAI, and Nvidia sells Microsoft a sweet, sweet stack of silicone to get the job done. Simple.

But who's the real winner? The end-user, of course!

Just kidding, the real winner's Nvidia, of course and as always. As Baron Harkonnen of Frank Herbert's Dune tells us: "He who controls the spice controls the universe." And Hopper's the spiciest spice in town, right now.

Well, it'll be Blackwell, soon, but that's Nvidia, too. Huang probably made the right choice going into tech and ditching a promising table-tennis career.

We all know that AI involves billions of dollars. One can't build data centres full of thousands of massive GPUs, pay the energy bills, and hire staff to program them for a mere pittance. But should that expenditure be used to create something for non-profit making or to create something that will ultimately rake in huge sums of money? When it comes to OpenAI, creators of ChatGPT, and Elon Musk, owner of xAI and Grok, it would seem they've not seen eye-to-eye on this matter.

OpenAI started as a non-profit AI research organisation in December 2015, before launching a for-profit subsidiary (OpenAI Global) four years later. The company was co-chaired by Sam Altman and Elon Musk until 2018 when Musk resigned from the board of directors, to avoid a conflict of interest with Tesla, which uses AI for its autonomous driving systems.

Over the years, other staff left OpenAI to start up AI companies of their own, and OpenAI Global attracted the attention of Microsoft, which pumped billions of dollars into the subsidiary, securing a nearly 50% share of the company.

Earlier this year, however, Elon Musk sued OpenAI and Sam Altman for breach of contract, violating fiduciary duty, and unfair business practices—in short, because OpenAI had decided to target profits and commercial interests, over the company's original goal of pursuing AI for the good of mankind.

Fast forward to a few days ago and it would seem that all is not quite as simple as it sounds, because OpenAI has published redacted DMs and emails that show Musk wanted OpenAI to go for-profit in the summer of 2017, before then going on to "demand majority equity, absolute control, and to be CEO of the for-profit."

In one email supposedly from Musk, he wrote that "the Preferred A investment round (supermajority me) should have the right to appoint four (not three) seats. I would not expect to appoint them immediately, but, like I said I would unequivocally have initial control of the company, but this will change quickly."

While it's not an outright demand for the control of OpenAI, I think it's fair to say that the message doesn't read particularly well and the board rejected Musk's option. It tried to placate him, though, saying "We really want to work with you. We believe that if we join forces, our chance of success in the mission is the greatest. Our upside is the highest. There is no doubt about that. Our desire to work with you is so great that we are happy to give up on the equity, personal control, make ourselves easily firable — whatever it takes to work with you."

Not that this had any effect because Musk's response was unequivocal: "Guys, I’ve had enough. This is the final straw. Either go do something on your own or continue with OpenAI as a non-profit. I will no longer fund OpenAI until you have made a firm commitment to stay or I’m just being a fool who is essentially providing free funding for you to create a startup. Discussions are over."

From the various messages being now made public, Musk was seemingly never hugely opposed to opening up a for-profit OpenAI offshoot or even necessarily for the whole company to go that way. The issue seems to be how it would all be organised and funded. Which does make his current lawsuit look maybe a bit spurious: why is there an issue with a for-profit OpenAI now?

Where things go now isn't clear, though I imagine that Elon Musk will have a few things to say on the matter. As far as I can tell, there's no mention of OpenAI's claims on his X account but the man posts a lot so there could be something buried away. The sensible thing to do would be to leave this all in the hands of legal beagles, but I guess OpenAI's post rather sidesteps that.

Time to get some more popcorn in, methinks.

Let's get down to brass tacks. Or beach balls, one of the two. (Stop it now - Ed). Deepfakes are a serious concern, with a recent Ofcom survey reporting that two in five participants said they'd seen at least one AI-generated deepfake in the last six months.

Deepfake content has the potential to harm and spread disinformation, so Meta releasing a tool to watermark AI-generated videos is probably a net benefit for the world (via TechCrunch). Meta Video Seal is open source, and designed to be integrated into existing software to apply imperceptible watermarks to AI-generated video clips.

Speaking to TechCrunch, Pierre Fernandez, an AI research scientist at Meta, said: "We developed Video Seal to provide a more effective video watermarking solution, particularly for detecting AI-generated videos and protecting originality.

"While other watermarking tools exist, they don’t offer sufficient robustness to video compression, which is very prevalent when sharing content through social platforms; weren’t efficient enough to run at scale; weren’t open or reproducible; or were derived from image watermarking, which is suboptimal for videos."

Meta has already released a non-video specific watermarking tool, Watermark Anything, and a tool specifically for audio, called—you guessed it—Audio Seal. This latest video-focussed effort is designed to be much more resilient than similar software from DeepMind and Microsoft, although Fernandez admits that heavy compression and significant edits may alter the watermarks or "render them unrecoverable."

Still, anything more resistant to removal than the current options strikes as a good thing, as a quick Google reveals multiple methods of AI-generated content watermark removal, which I won't link to here. Still, the info is out there, and that means potentially harmful content is currently going unchecked.

With a watermark in place, news outlets and fact checkers the world over can correctly determine whether a video is potentially real, or created by AI. The real uphill battle now is widespread adoption, as no matter how effective the tool is at flagging AI videos, it'll matter not a jot if no-one uses it.

To that end, Meta also has plans to launch a public leaderboard, called the Meta Omni Seal Bench, which will compare the performance of watermarking methods, and is currently organising workshops at next years ICLR AI conference. Fernandez says the team hopes more AI researchers and developers will integrate watermarking into their work, and that they want to collaborate with the industry and academic community to "progress faster in the field."

So come for the seals, stay for the fight against AI-generated misinformation. Sometimes it can feel like we're drowning under a wave of an AI industry moving incredibly quickly into the future—with minimal checks and balances—so it's difficult not to applaud Meta for making attempts to rein in some of the more egregious side effects.

Particularly if it makes me think of this fella. You're welcome.

An interesting concept, no doubt. However, there's one fly in the ointment: Even huge titans like Google struggle to produce AI-generated responses that make any sense. 404 media has been speaking to Clint Basinger, a YouTuber who's been testing out "editable AI-enhanced reply suggestions" on the platform—and according to Basinger, the results so far have been variable at best.

The suggestions work in roughly the same way Gmail creates optional "smart reply" suggestions to your emails, except more in-depth. The AI responses appear to be based on actual comments from the creator in an attempt to mimic their tone, which on paper sounds like a perfectly reasonable idea. Context is everything, however, and without sufficient contextual data, the AI seems prone to making things up.

In a video demonstrating a Duke Nukem branded energy drink powder (god help us all), Basinger struggled to find a scoop inside the packaging. A commenter suggested that the scoop might be buried in the powder, to which the AI suggested the response:

"It’s not lost, they just haven’t released the scoop yet. It’s coming soon."

Later on in the video, Basinger shakes the container, and a commenter suggests that he should have had a tighter grip on the lid. According to the YouTuber, the AI suggested they respond with: "I’ve got a whole video on lid safety coming soon, so you don’t have to worry!"

Mind you, I suppose you could argue it's providing ideas for new content, at the very least. Lid safety videos are the hot new thing, after all, and... I kid, I kid.

It's not just context the AI seems to struggle with. Occasionally, it's repeating words. Basinger's other channel is called LGR Blerbs, and upon viewing a new video, another commenter said "Nice. Back to the blerbs."

"It's a whole new kind of blerp," suggested the AI.

Yep, a completely made-up word there. Anyway, Basinger seems to take all this in good humour, although they report one incident where things take a slightly darker turn. When another commenter expressed delight that Basinger was posting to their second channel, the AI suggested they respond:

"Yeah, I'm a little burnt out on the super-high-tech stuff so it was refreshing to work on something a little simpler."

While YouTube does provide advice for its creators in regards to burnout, I doubt it'd want its AI to reference the fact that creators working on the platform can experience high degrees of stress and strain keeping up with the churn.

So, as for that dead internet theory? It'd be particularly obvious with this bot I think, although that's not to say more advanced ones (or perhaps even worse ones, given some of the horrors seen in the YouTube comment section) aren't out there, auto-generating away.

And as a useful tool for creators? I can't quite see the idea of AI-generated responses catching on. After all, aren't you, the audience, there to listen to what your creator of choice has to say, not what the AI thinks they might say?

And goodness knows what the auto-comments would look like with a more controversial training subject. I shudder at the thought.

Still, I digress. Time to cool my circuits, and... oh no, dear reader. The game is up. You're really here though, aren't you? Right? Please tell me I've been writing this article for someone, at the very least.

The Mozilla Do Not Track support page states (via The Register): "Starting in Firefox version 135, the 'Do Not Track' checkbox will be removed. Many sites do not respect this indication of a person's privacy preferences, and, in some cases, it can reduce privacy.

"If you wish to ask websites to respect your privacy, you can use the 'Tell websites not to sell or share my data' setting. This option is built on top of the Global Privacy Control (GPC). GPC is respected by increasing numbers of sites and enforced with legislation in some regions."

This might initially sound like not such a bad thing, the way Mozilla talks about it. But it is, in my opinion, part of a broader trend to bait and switch general privacy concerns for more specific ones. Let me explain.

DNT is a request header that asks sites you visit—you guessed it—not to track you. Websites can then decide whether to adhere to this request, but the idea is to make it so users can easily signal to sites their privacy preferences without having to set these preferences for every site. Whether sites have to adhere to these requests would then be a legal matter depending on the laws in different regions and so on.

While it's true that most sites simply ignore these requests, I'd argue that's a legal or enforcement issue and not an issue with the DNT request specification itself. This is in the same way that there's nothing wrong with requesting people don't punch you in the face. Even if people keep ignoring that request and get away with it, the request itself is reasonable, don't you think?

The argument, or at least the implication, seems to be that we shouldn't worry because Global Privacy Control (GPC) is the new replacement for DNT, and it's better respected by websites and sometimes actually enforced.

This might be true, but seemingly buried in the small print is the crucial fact that GPC doesn't as sites to stop tracking you like DNT does. It asks them to stop selling the data that it does track. Its specification refers to "do-not-sell-or-share" interactions or preferences, not do-not-track ones. Bait and switch, much?

This is, of course, better than nothing. And it's perfectly fine for those who were only concerned about their data being sold. But I'd bet that at least some users who were keen on DNT didn't want their data to be tracked at all, in principle—at least not by default. It's not just that users might want privacy in the relation between themselves and the site in question, protecting information from outside sources. It's that they might want privacy full-stop, including from whichever website they're visiting.

Plus, opening the door to "tracking, but not selling" could still mean companies you aren't aware of accessing your data, because a website might not sell your data but might give it to a partnered company, for example. The corporate world is wild, and you can bet if there's a way around things, some companies will find it.

It's not as if DNT was in principle unenforceable, either. Only last year a German court ruled that LinkedIn had to listen to DNT requests.

Whether the trend towards swapping out general privacy concerns for more provincial ones is a sign of people throwing in the towel on general privacy because it's difficult to enforce, or whether it's people and companies actively deciding to allow companies to continue to guzzle our data, it doesn't matter. The result is a continued normalisation of privacy erosion.

Of course, Mozilla might not intend any of this, and it might just be a "well, what's the point anyway?" response to DNT not being taken up by websites and courts at large. But I can't help but wonder: Why bother removing the setting instead of having it as an additional option? GPC and DNT request headers could both exist side-by-side.

One might argue that a DNT setting could mislead users into thinking that their activities aren't being tracked when actually it depends on the website adhering to the request. But surely that's something a simple warning could fix. And at any rate, the same would be true of GPC.

Mozilla was the first to implement DNT, so it'll be particularly sad to see the option disappear from Firefox for that reason, too. Here's hoping something better comes along, something which is legally binding and easily enforceable. I won't hold my breath, though.

Central to the update is the new "Home experience," previously only available to Xbox Insiders, that will help players "stay up to date on game releases, new events, Game Pass content, sales, and more," Xbox general manager of content curation and programs Chris Charla wrote (via The Verge). "There are also new curated collections and recommendations, deals and discounts, and the Jump back in feature, which enables players in Compact Mode to click on any game card and jump right back into gameplay from the game’s hub."

As for the new games added as part of the update, Charla said Microsoft began working with partners in the summer "to bring all PC games with Xbox features into the Xbox app."

"We're super excited to see what Xbox games come to Windows PC from game creators in the future and to welcome nearly 400 titles that previously weren’t discoverable or purchasable in the Xbox app," he wrote. "Among this list are dozens of games from veteran Japanese studios Kemco and Kairosoft, as well as games such as Shovel Knight Treasure Trove by Yacht Club the Jackbox Party Pack, The Invincible by Double 11 and many, many more."

I can't do a before-and-after comparison because I discovered earlier today that I didn't actually have the Xbox app installed on my PC at all. I do now, and it looks perfectly serviceable (if a little sluggish, although could be my low-quality internet rearing its ugly head again), but I think that really speaks to the challenge Microsoft is facing as it looks to carve out a bigger space in PC gaming: The Xbox app is so inessential that I didn't even know I didn't have it installed.

Not everyone is going to be in that boat, of course. Game Pass subscribers in particular may like the way it sorts and highlights current and upcoming games in various ways, and among the new games added are more than 100 Xbox Play Anywhere titles, with "universal Xbox ownership"—if you buy it on PC you also get the console version, and vice versa—"as well as universal cloud saves on Xbox, and cross-play between Xbox versions on console and PC (and other platforms, at the developer’s discretion)." For players invested in that PC-console cross-compatibility, this update will be a big improvement.

But Microsoft has greater ambitions than merely making the Xbox app a more welcoming home for Game Pass subscribers: "While many players use the app to find and launch awesome titles from Game Pass, the Xbox app is also becoming a great way for players to find cool Xbox games that aren’t available with Game Pass," Charla wrote.

Microsoft will be testing additional new features for the Xbox app over the next few months, beginning with a small collection of casual games including Candy Crush Soda Saga and Angry Birds 2 later this month. "This work is always done with the goal of delivering players the best and most diverse array of games possible and providing game creators with access to new discovery surfaces and new audiences," Charla wrote.

Charla said Microsoft will share more about its plans for the future of the Xbox app at GDC, which takes place March 17-21, 2025.

As reported by Sweclockers, the PC gaming platform managed to hit a total of 39,205,447 players on Sunday, December 8 at 2 PM UTC, as can be stopped on SteamDB. 12,090,313 of those were actually in games, with the rest having the Steam app open to some capacity, casually browsing, downloading games, or looking through the store.

Notably, this huge number is partially influenced by the dual launch of both Marvel Rivals and Path of Exile 2, with a few single-player releases like Dragon Age: The Veilguard, Black Myth: Wukong, and Baldur's Gate 3 contributing tens of thousands

Marvel Rivals, the superhero universe’s attempt at the Overwatch formula is not only free to play but hotly anticipated thanks to recognition of the name and a fun cast of characters. That's sure to have significantly helped the player count. At the record-breaking moment, 368,021 players were in-game.

Path of Exile 2 is an even more impressive launch as it's a paid (albeit fairly cheap) early-access launch. The game has been doing so well that its devs haven't been able to fully keep up with the demand on its servers. There were 470,715 players in this action RPG at the moment of the record, and the player figures could have been even higher if server issues didn't persist over the weekend.

This concurrent player figure is even more impressive when you consider that, in January of this year, the previous record of 33.5 million players was beaten by 33.6 million at one time. Then, in August, Steam managed to surpass 37 million concurrent users, thanks to the rather explosive launch of Black Myth: Wukong with 2.4 million concurrent users by itself. This placed it in second behind PlayerUnknown's Battlegrounds, as the game with the highest all-time peak on Steam.

The launch of Marvel Rivals and Path of Exile 2 only account for just shy of a million of this new record so this suggests a general uptick in many games, not just a handful. That weekend saw big player figures in Counter-Strike 2, PlayerUnknown's Battlegrounds, Dota 2, and Grand Theft Auto V too.

As we reach the end of the year, more players seem to reaching for video games. When the weather is as dark and gloomy as it is here in the UK, who can blame them?

Despite rapid advances, the accuracy of generative AI in making images and videos is still far off the mark at times, but voice reproduction is already remarkably good. With the use of vocal cloning on the rise, to carry out scams and fraudulent claims, the US Federal Bureau of Investigation has issued a list of tips (via Ars Technica) to help protect yourself, including the advice that you should create a secret word or phrase that only you and your family know.

Vocal cloning is a process where various audio clips of a person speaking are used to train a generative AI model so that it can then be used to replicate that person's normal speech patterns, tone, timbre, and so on. For example, it was used to create Google's 'podcast hosts' for its NotebookLM system and one would be hard-pressed to identify that it's not real people speaking when you listen to it.

One doesn't need to be an expert in AI to see how such a thing can be misused for nefarious purposes. And even though one would need genuine clips of your voice and speech mannerisms to clone you, it does mean there is a chance that someone out there could attempt to use 'you' in order to carry off a scam of some kind.

This is where proof of humanity comes in—essentially it's a password or phrase between you and your family, or more accurately, an MFA (multi-factor authentication) system where your voice is one factor and the password is another. A generative AI system is most likely going to be stumped by the bizarre and non-sensical because it wouldn't fall in line with the usual lines of predictability when it comes to talking.

Asking "What does the donkey say to the aardvark?" might seem like the opening line to a joke but if you're expecting "Wednesday afternoon, in Nepal" as a reply, few generative systems are going to offer that reply. So when you don't get that response, you can be immediately suspicious as to the genuineness of the call.

It's not a foolproof system to catch an AI scam, as it does require all parties involved to remember the password, and the correct response, and be able to do it at a moment's notice. If my mother suddenly asked me about donkeys and aardvarks out of the blue, I think my first response would be 'Err…what?' and then the phone would get slammed down, every time.

And then there are the situations where it's not a family member on the phone, but the boss or your manager of the company you work for. Is one expected to have a unique password for everyone there, to ensure the business never falls for a scam?

AI voice cloning is arguably a natural extension of the usual email scam, where a higher-up appears to be asking you to send them money or sensitive information—if that person is publicly well-known and clips of them talking can be sourced, such scams would be relatively easy to carry out.

I don't think there's any perfect solution to counter any AI voice scam, unfortunately. Even if the top companies in artificial intelligence, such as OpenAI, Google, and Microsoft integrate systems that always make it obvious that it's an AI 'talking' and not a real person, there will always be people out there who are able to create a similar model that doesn't.

Hopefully, smarter folks than me are working on solving this thorny problem.

Okay, perhaps not quite this pairing, but a team up between a top analytics outfit and an autonomous defence systems vendor has real-world implications that might be worth paying attention to regardless. Palantir Technologies specialises in software platforms for big data analytics, while Anduril is a US defence technology company with the stated mission of "transforming defence capabilities with advanced technology."

Together they've announced a new consortium to "ensure that the US government leads the world in artificial intelligence" (via The Register). Exactly what that means is hidden among an impressive array of corpo-speak, but there's some language here that instinctively makes the hairs stand up on the back of my neck.

"We will utilize Anduril’s Lattice software system and the Anduril Menace family of deployable compute and communications systems to instrument the tactical edge for the government’s secure, large-scale data retention and distribution.

"Lattice connects directly with third-party defense systems at the edge, delivers autonomy to machine operations, securely distributes their information across a large-scale data mesh, and backhauls all tactical data into government enclaves for the purposes of AI training and inferencing.

Menace devices are also purpose-built for the tactical edge, customized down to the silicon level for the unique requirements of national security operations in tactical environments—including, soon, next-generation encryption."

With the best will in the world, calling one of your systems "Menace", as a defence company, doesn't strike as particularly great optics. Anyway, the gist of the partnership is that there are supposedly two limiting factors to adopting AI for national security purposes: Data readiness, and secure enterprise pipelines to "turn that data into AI capabilities."

To solve this, Anduril's systems will connect to third-party defence systems and gather data, before Palantir's AI platform structures and labels it for AI training, while also providing a secure pipeline to deploy AI models into national security systems. That's a lot of "systems", I know, but parsing and simplifying this sort of doublespeak seems like the kind of job that a machine learning model would actually be better at than me.

In the Tolkienverse, a palantír is an indestructible crystal ball used for communication and to see events in the past, the same kind that cheeky hobbit Poppin uses to gaze into the eye of Soorun (I'll stop it now) before Gandalf snatches it away from him and gives an appropriate scolding.

Andúril, meanwhile, is the name of Aragorn's gigantic sword, reforged from the shards of Narsil and also referred to as the "Flame of the West."

Which, given the context, seems somewhat appropriate. Or slightly terrifying, you pick. Anyway, this is what happens when grown nerds start naming companies after objects from their favourite universes, so I look forward to writing about future team-ups between Death Star Enterprises, Phaser Corp., and the Infinity Stone Institute. What a world, eh?

Oh, and if I've got any of the LOTR lore here wrong, feel free to write in to any of my editors at... actually, on second thoughts, don't. It probably wouldn't be appreciated.

Microsoft has confirmed that Ubisoft's temporary patches for Avatar: Frontiers of Pandora and Star Wars Outlaws resolve problems with crashes, though you still might experience performance issues. That means if you have either of those games installed (fully patched, of course) then you'll eventually be offered the Windows 11 24H2 update, should you be using that operating system.

However, if you have Assassin's Creed Origins, Odyssey, or Valhalla installed, then Microsoft's big update for Windows will still be unavailable and you're strongly advised not to circumvent the block by installing Windows 11 24H2 directly (e.g. via Microsoft's media creation tool).

The big three AC games all use Ubisoft's proprietary Anvil engine. Avatar and Outlaws, thought, use Snowdrop and while that's also a Ubisoft engine, it's predominantly developed by Massive Entertainment. Microsoft hasn't given any indication as to what's causing the problems with 24H2 and these specific games, but the fact that the Snowdrop-powered ones have been fixed first clearly suggests that it's something fundamental with how the engine interacts with the operating system.

It's such an unusually specific issue, with no other games that I'm aware of having the same problems with Windows 11 24H2 (freezing during loading or gameplay, crashing to desktop, black screens), that this might not be Microsoft's fault at all. Or at the very least, it's not something that its developers have messed up. At the same time, I don't believe that Ubisoft is at fault either as all the affected games work perfectly fine with 23H2.

Although the 24H2 update does have some nice features to it, there's nothing there that really demands one should install it at all costs, so if you are a big fan of Assassin's Creed Odyssey like me and you were planning on installing it over the Christmas break to enjoy some long hours exploring Greece, then double-check what version of Windows 11 you have before you do.

Just head to the Settings option in the Start Menu, then System and scroll down to About. If you already have Windows 11 24H2, then you'll either have to forgo Origins, Odyssey, or Valhalla until they're patched or switch back to using 23H2.

That will involve reinstalling Windows if it's been more than 10 days since the 24H2 update was installed, but if that's happened recently, you can use the 'roll back' feature. Head to Settings, then Windows Update, Advanced Options, Recovery, and then you should see a button that says 'Go Back.'

Probably best just waiting for Ubisoft to patch its games, rather than go through all that malarky, if you ask me.

On an average day about a dozen new games are released on Steam. And while we think that's a good thing, it can be understandably hard to keep up with. Potentially exciting gems are sure to be lost in the deluge of new things to play unless you sort through every single game that is released on Steam. So that’s exactly what we’ve done. If nothing catches your fancy this week, we've gathered the best PC games you can play right now and a running list of the 2024 games that are launching this year.

Granny Escape Together

Steam ‌page‌ ‌
Release:‌ December 6
Developer:‌ CreepyToad

Steam received more new indie horror games than usual last week—and it usually gets a lot—but I couldn't turn down the opportunity to highlight this co-op game about escaping a crazed grandma. The set up is simple: after encountering a weird dancing elderly woman in the forest, you receive a surprise blow to the head and awaken later in the decrepit abode of granny. What follows is the usual interleaving of hide and seek with tense puzzle solving, as you and up to three friends work hard towards one of three escape routes. I've never met a granny I didn't like, but the granny in Granny Escape Together looks utterly terrifying. If you dare, it can also be played solo.

Nikoderiko: The Magical World

Steam‌ ‌page‌
Release:‌ December 6
Developer:‌ Knights Peak

This lush 2.5D platformer is strongly reminiscent of Retro Studios' Donkey Kong offerings. It also evokes the cruisy atmosphere of '90s media, with its nostalgic anthromorphs and David Wise composed soundtrack (in case he needs introduction, think Banjo Kazooie and Diddy Kong Racing). The Donkey Kong comparisons are definitely invited: there are minecart rides here, the platforming is deliberately tricky, and couch co-op is supported, so a friend can shout at you every time you fall into a bottomless pit. If you've already played Kaze and the Wild Masks (a brilliant and underrated action platformer), this looks like a good follow up.

Rings of Zilfin

Steam‌ ‌page‌ ‌
Release:‌ December 6
Developers:‌ Strategic Simulations

At first I thought this was a new retro-styled adventure in the vein of something like Skald, but Rings of Zilfin actually released for the Commodore 64 in 1986, so it's the real deal. This re-release—which hit GOG back in April—doesn't modernize the game at all, so you should go in expecting a typically cryptic 1980s RPG with no hand-holding. That said, this enlightening post on crpgaddict insists that it's relatively easy by CRPG standards, so if you're going it just to enjoy the gorgeous '80s pixel art and the vibe—which is what appeals to me—then you should be OK.

The Hole

Steam‌ ‌page‌ ‌
Release:‌ December 6
Developer:‌ Bober Bros

Bober Bros follows up its disturbing It's Just A Prank with yet another grimy, dubious-looking horror game that makes something like Murder House look cosy. Played from a first-person perspective, you're a habitual drinker watching TV in your apartment, minding your own business, until some distracting drilling starts next door. Soon, holes begin to appear in your wall. I'm not sure what happens from there, but expect 40 minutes of superlative discomfort.

Esca

Steam‌ ‌page‌ ‌
Release:‌ December 7
Developer:‌ LuN

Esca is a murky, atmospheric survival RPG played from a top down perspective. As a maintenance robot, you'll navigate a mysterious underwater research station, though by the looks of it this station is basically destroyed beyond repair. As the trailer above makes clear, this is borderline a horror game, but it's a quiet and ruminative one, with branching narrative paths and a molasses-thick mood of dread.

Well, at least it's not calling me 'a stain on the universe' like Gemini AI allegedly told one user recently. Whereas Genie 1, revealed back in February, could only cobble together 2D scenes, the just announced Genie 2 is a step up, offering somewhat explorable 3D game environments (via PC World). I say 'somewhat' because it isn't long before the wheels come off.

For one thing, the player character model struggles to look consistent throughout movement. For another, playable worlds generated by Genie 2 don't last long; Google's writes, "Genie 2 can generate consistent worlds for up to a minute," though admits that most of the examples they show lasted between 10 and 20 seconds. So, no, you won't feed Genie 2 a prompt regarding that long awaited sequel and be satisfied by the results.

Google's prompting process is also not as straightforward as typing, "One Cyberpunk game, please," and diving into a Night City knockoff either. Genie 2 generates its game worlds in response to still images that were themselves generated by Imagen 3, a text-to-image model also from Google. That's a whole lot of snakes eating their own tail.

Genie 2 itself is a "an autoregressive latent diffusion model" that draws from "a large-scale video dataset" that Google doesn't really detail the content or source thereof otherwise. Google has also experimented with feeding photos of the real world to Genie 2, highlighting responses that "model grass blowing in the wind or water flowing in a river." The Gifs they share of these responses are about as muddy looking as you'd expect, though.

Credit where some credit may be due though, Genie 2 does make strides in a number of things generative models often struggle with. While the player character model warps and blurs like a water damaged printout, the environments one can trundle through remain surprisingly consistent—definitely bland and generic, but consistently so.

For instance, Genie 2's "Long Horizon Memory" remembers aspects of the environment after they disappear from view. Say you're running between pyramids—when you look back, the pyramids will stand in the same spot you last saw them, rather than teleporting around behind you like it's a game of 'What's the time, Mr. Wolf?' Similarly, Google touts Genie 2's 'counterfactual' capabilities, which is really just a fancy way of saying that multiple players can play the same generated level and have a consistent experience.

Google has shared a mix of examples controlled by humans and SIMA, their own AI player "designed to complete tasks in a range of 3D game worlds by following natural-language instructions." Many of the shared excerpts show digital avatars controlled with typical WASD keyboard controls. Google writes, "Genie 2 responds intelligently to actions taken by pressing keys on a keyboard, identifying the character and moving it correctly. For example, our model has to figure out that arrow keys should move the robot and not the trees or clouds." Good for you, Genie 2.

But besides walking around these occasionally wonky looking levels, Genie 2 can also generate object interactions and even NPCs. Naturally, the NPCs have nothing of note to say, but Genie 2 can generate balloons to burst and barrels to explode—though that's hardly the most compelling gameplay loop.

While no one outside of Google can yet play around with Genie 2's output, the company is eager to tout its potential use cases, such as rapid prototyping based off of concept art. This may sound appealing—right up until you ask what happens when a developer wants to slightly adjust absolutely anything about their AI generated prototype.

While Genie 2's snatches of game may spark the imagination of AI-defenders, I'm far from convinced—and honestly, I'm just a wee bit concerned about its potential labour implications for game development. Earlier this year, Take-Two CEO Strauss Zelnick dismissed the claim that AI would take away jobs, arguing, "It's not going to make people irrelevant. It's going to change the nature of certain forms of employment. And that's a good thing." But as Andy Chalk points out in his news piece, industry transitioning to large scale automation is not always a painless process. It's my sincere hope that Google's little experiment doesn't end up motivating even more layoffs.

Warcraft 1 and 2 will leave GOG this December 13, exactly one month after Blizzard surprise-dropped its remasters of the games, but if you already own the games—or buy them before the 13th—you won't lose access. GOG is actually launching a discount code, MakeWarcraftLiveForever, that you can use to knock $2 (£1.50) off the Warcraft 1 and 2 bundle if you pick it up before D-Day (the D is for delisting).

There are a couple of wrinkles, though. One, Blizzard also sells the OG Warcrafts on Battle.net, and there's no word yet about whether they're also being delisted there. I've reached out to ask about that, and I'll update if I hear back. Even if Blizzard does keep selling them, though, there are some differences between the Battle.net versions and GOG's. For instance, the GOG version of Warcraft 1 uses an internal config tool to make multiplayer connection a little easier, while GOG's Warcraft 2 comes with a new DXWrapper that tries to capture the game's original presentation on modern resolutions.

The second, bigger wrinkle is that Warcraft 1 and 2 were only just inducted into the GOG Preservation Program, the store's scheme to use its own devs to keep a select list of golden oldies running on modern hardware "for generations to come." What does it mean for GOG's pledge if a game on its list is no longer purchasable on its store?

Well, props to GOG: it's promising to keep maintaining Warcraft in perpetuity for everyone who picked it up before the delisting. "Once the game obtains a badge, we pledge to maintain its compatibility even if it gets delisted from the store," it says, "This means that owners of those titles can still expect a seamless experience and tech support for those titles," up to and including dealing with changes to Windows that muck with old games' playability.

So, to be clear, that's not just an announcement about Warcraft 1 and 2, but a commitment to a whole policy on delisted games. Whether you can buy 'em or not, once a game's on GOG's preservation list, it stays on the preservation list. It's quite a commitment to make to products that aren't going to directly make you any money (although the boon to GOG's reputation probably has its own indirect rewards), and it's good to see. I suspect keeping Warcraft 1 and 2 playable won't be too laborious for devs, but I wonder what the long-term ramifications of the policy will be if more publishers choose to yank older, cheaper versions of games from sale. Still, so long as we have Alpha Protocol, I think we'll be okay.

Windows 11 has a dominant position as the de-facto default operating system for PCs. Buy a new laptop or desktop and there's a 99% chance it will come with Windows. There are a few exceptions, such as the Steam Deck's Linux-based SteamOS, or the small number of machines you can buy without an OS installed, but pick up a gaming PC from any of the major players and you're getting Windows 11.

It's also where the games are. The number of items on Steam tagged as native Windows-compatible is 207,105 at the time of writing, with about 50,000 for macOS and somewhere in the region of 35,000 for SteamOS/Linux. Some of these are going to be DLC cosmetics or mission packs, but there's still a lot more availability on Microsoft's platform than for the others.

Windows 11 picked up a bad rep at launch. Some of this was down to aesthetics, with the moving of the Start button and taskbar icons to the centre (easily fixed in the Settings app) coming in for particular ire. The Settings app has also been under fire, with detractors asking why it exists, when there was already a perfectly good Control Panel, and noting that splitting settings between two locations is confusing. There's also a perceived lack of customisation ability and settings: take the taskbar, which can be placed in the centre or to the left of the bottom of the screen, but not at the top or side of the desktop in the way some perverts like it. These are, it has to be said, all good points, but they don't really affect the OS's gaming capabilities.

Then there's the hardware security requirement, which prevents the OS from running (easily) on PCs not made in the last seven years or so. People seem oddly attached to their old hardware, but the benefits of secure boot for malware prevention are worth the minor inconvenience. Was there a similar outcry when Windows XP introduced Windows NT-like account security to a consumer OS? We're much too young to remember. Cough.

Even earlier than XP, Microsoft tried to reinvent Windows 98 around the web, and it didn't go well. Buried in that OS, and still there in Windows 11 (Folder Options, General tab), is the ability to select a file or folder by hovering the pointer over it, with a single click to open as you would with a hyperlink in a web page. No one has ever used it, but Microsoft's willingness to experiment with new ways of interacting with PCs has to be applauded even if it does result in Windows 8's fullscreen Start menu and Metro apps, an innovation so poorly received that there was no Windows 9 in an attempt to dig a moat around the idea.

So Windows 11 makes people feel sad and is hard to ignore. But is it actually bad?

Windows 11 (and Windows 10, to an extent) sends information about your activities back to Microsoft, which some claim spikes CPU usage. Then there are the ads—you're asked during initial setup of a new Windows 11 device whether you want adverts to be tailored to you or more generic, with the caveat that the number of ads you see won't change (hint: turn off Widgets in Taskbar Settings). Microsoft Recall, the AI-catalogued searchable snapshots that Copilot+ PCs will take of your usage so you can more easily find something you half remember googling three weeks ago (and which seems very much like a solution in search of a problem) is being criticised as a privacy-invading nightmare and destined to be the new most quickly-switched-off feature.

These can be seen as objectively bad things, but the fact remains Windows 11 is stable, keeps itself and its drivers up-to-date, boots up quickly and, on modern Intel processors, has a core scheduler that can more effectively spread the workload between P and E cores. It also adds Auto HDR and Dynamic Lighting and makes games easier to run in windows (er, windowed). It's pretty dual-boot friendly, so there's nothing to stop you from having another OS installed to test out, diving back into the familiar clutches of Microsoft if it all gets a bit too much.

As for performance, things look very close. PC Gamer's sister site Tom's Hardware installed Windows on a Steam Deck and wrote down some frame rate comparisons which see Windows a little bit ahead in some games, and SteamOS take the lead in others. Windows on Deck isn't a perfect experience, as our man Dave found out, but SteamOS is a release that's tailored to its specific hardware, like the custom version of FreeBSD that the PlayStation 5 runs, and is only available to download as a recovery image and not a flawless experience off-Deck. It's possible to install GPU drivers and even Valve's Proton abstraction layer on Linux (and some of them can be very good, as Tom's Hardware discovered), but you don't get the SteamOS frontend and Big Picture Mode can be finicky.

However, you can do it if you want, and that's always been the guiding light of PC use. You don't have to be tied in to one manufacturer or methodology, which is part of what makes the PC the best gaming platform as well as all the other things it can do (spreadsheets? No idea). Windows 11 has been available for three whole years now (though oddly it feels longer), and thanks to the constant stream of patches and feature updates has matured into an excellent place to game. Any enterprise of this complexity is going to come with bugs, flaws, and things that make you screw up your face in bewilderment, but there's no denying it's popular.

In the October 2024 results of the Steam Hardware Survey, Windows 11 has overtaken Windows 10 to gain 48.8% of users. Win 10 sits on 47.46% (and is slowly falling), for a total Windows market share of 96.54%—including the 0.28% of gamers still using Windows 7. Linux (mostly Arch as it's the tech behind SteamOS, plus Ubuntu) sits on 2% and macOS on 1.39%. There's no arguing with those figures, and the bulk of development efforts from game companies, hardware driver writers and other software providers is going to be focused on Windows 11, love it or hate it, because that's where the audience is.

The deal with Epic will leave ownership and full creative control over the new game with Section 9. Once the game is released and costs are recovered, the studio will earn at least 50% revenue share on sales.

"Working with the Epic Games Publishing team has been an amazing experience for us," Section 9 Interactive co-founder Gustaf Heinerwall said. "Ever since our first meeting with the team at EGP [Epic Games Publishing] we felt that this partnership was a perfect fit for our studio. They’ve given us incredible support and the freedom to fully explore our creative vision, which is all we can ask for in a publisher. We are thankful for this opportunity, and excited for what comes next."

"What comes next" will hopefully be a proper look at the new game. It was revealed nearly four years ago, in January 2021, and we've seen nothing of it since aside from a "we are hiring" message posted last year. Today's post about the Epic publishing partnership is the first activity on Section 9's X account since August 2021, when the studio announced that it was an Epic MegaGrants recipient. Epic hinted at more to come in the hopefully-not-too-distant future in its own post on X: "They’re DEEP in development… stay tuned for updates."

More and more, I find myself thinking about that New York Times article, We Should All Know Less About Each Other. Okay, yes, that was a far more serious insight into an increasingly interconnected world and one research paper's (failed) attempt at breaking down political echo chambers. Discord's all-seeing gamer eye isn't quite so crucial to our society, but it sure is starting to feel like a lot.

Seeds of Discord

Now listen, I know the concept of online statuses is older than some of you reading this. I was insistent on making sure everyone knew which emo banger I was giving myself hearing damage with on MSN, and seeing my high school boyfriend in the trenches of Nuketown on the Xbox 360 was usually a pretty good indicator of why he wasn't texting me back on BBM.

They've literally been around for as long as I can remember at this point, and the thing is I do like them! I like seeing when my friends are playing a game I also want to play, almost like an unspoken invitation to dive in and say hello. It's cool to see someone else playing something I like but haven't seen too many others trying out, giving me the perfect conversation starter I may not have otherwise had without that knowledge.

But has anyone else noticed Discord is starting to act like a bit of an overbearing friend, keeping tabs on what we're doing and what we have been doing like, all the freakin' time? Like okay Discord, I was cool with you telling everyone what I was playing. I didn't even really mind your Rich Presence integration for some games, giving my friends a lil' extra info on what exactly I was getting up to within whatever game I was playing.

What I'm not all that cool with is the whole Recent Activity thing that got added a few months back basically giving a play-by-play on how I spent my last several days gaming. I did not need my friends to know I'm playing Overwatch 2 for the first time in 10 months—they bully me enough for my life choices as is—nor did they need to know I am on my millionth streak of Final Fantasy 14 despite the fact I keep whinging that I have nothing to do in the game.

Oversharing is not caring

Yep, Discord is a proper gossip right now! It tells damn near anyone who'll listen when you last played something, how long you played it for, how many days in a row, and the number of hours you've dumped into it this week, you wee sicko. Almost every time I enter a server it's telling me which of my friends desperately need to touch grass, unless the feature has been dutifully disabled by a moderator.

Sure, I can just turn it off, but with it I have to go totally dark and entirely hide what I'm playing. If I'd rather not have everyone know I'm on my 20-day sicko-streak for my latest gacha game obsession, then nobody can know I'm playing it at all. I'm either forced to walk onto the stage stark naked for all to see, or I don't get to perform Macbeth at all. And I at least want people to know I'm performing Macbeth, you know?

I do wish I had a little more agency over how much of an oversharer I can be on Discord. I use it every single day, it's where I talk to almost all of my friends and it's even a good metric for figuring out what the Hot Game of the moment is, which unsurprisingly comes in handy for this ol' job. I don't want to be a total recluse, but I don't exactly want to be baring all, either. Just back off a little bit and give me some more personal space please, Discord, okay? Thanks.

"As part of an agreement that includes equity with milestones as a majority of its compensation, Dr Disrespect will provide exclusive content to Rumble Premium for his 'Champions Club' community," Rumble said. "Dr Disrespect will also lead Rumble Gaming, acting as an advisor and helping to build the Rumble Gaming community."

Along with exclusive content for Rumble subscribers, Beahm will continue to stream free content on Rumble.

The move to Rumble comes after Beahm's request to remonetize his YouTube channel, which was demonetized in June following his admission of inappropriate conversations with a minor on Twitch, was declined in September. Ironically, just a few weeks later he shot down rumors that he was moving to the streaming platform Kick with a disparaging post saying he'd retire first: "My community is the best in the industry," he said at the time. "I wouldn't take it there."

Which is fair enough: Kick is the platform that controversially pays huge amounts of money to xQc and Nickmercs for gambling streams, and more recently hosted banned Twitch streamer Adin Ross as he fawned over Donald Trump. But Rumble is home to numerous personalities of dubious repute too: Streamers currently trending on the platform include Dan Bongino, Steven Crowder, Viva Frei, Russell Brand, The Quartering, and the Tate Brothers. Is that worse? I would argue that it absolutely might be.

Beahm isn't a bad fit for it, though. In October he announced a line of Donald Trump-inspired "Make Gaming Great Again" merchandise available for purchase on his website. Naturally, it sold out quickly.

"Crazy to think a type of 'class of people' tried to cancel us and create a false, exaggerated narrative about something from almost a decade ago. They embarrassed my family and I," Beahm said on his real-name X account, which had been idle for more than a year prior to today. That "narrative" arose from Beahm's admission in June in which he downplayed his interactions with a minor that got him a lifetime Twitch ban, claiming they were "casual, mutual conversations that sometimes leaned too much in the direction of being inappropriate." Logs of the conversations have not been released by Beahm or Twitch.

"And with all that false, negative energy, YouTube decided to act as well and take my livelihood away by demonetizing us without any real explanation ... Truly grateful for the opportunity to help grow a new, hungry platform that represents the type of values I'm aligned with. No more cancel culture, no more suppression."

"Rumble is in a new era, and I'm laser-focused on expanding into two categories: gaming and crypto," Rumble Chairman and CEO Chris Pavlovski said. "Dr Disrespect will give a much-needed injection into our gaming category."

Beahm's first stream on Rumble is set to happen on December 2.

On an average day about a dozen new games are released on Steam. And while we think that's a good thing, it can be understandably hard to keep up with. Potentially exciting gems are sure to be lost in the deluge of new things to play unless you sort through every single game that is released on Steam. So that’s exactly what we’ve done. If nothing catches your fancy this week, we've gathered the best PC games you can play right now and a running list of the 2024 games that are launching this year.

Follow the Meaning

Steam ‌page‌ ‌
Release:‌ November 22
Developer:‌ Hiking Mind Studio

This bonkers point 'n' click adventure follows the affairs of detective Paul Trillby, who has arrived in a strange town to investigate a mysterious hospital. This is no average hospital, because its patients come back out without their memories. What follows is the usual mix of pointing, clicking, exploration and puzzle solving, all couched in an oblique, Kafka-esque atmosphere. It's inspired by the Rusty Lake series and Samorost, and runs for around an hour and a half.

Threshold

Steam‌ ‌page‌
Release:‌ November 20
Developer:‌ Julien Eveillé

While it looks like yet another PS1-styled horror game, Threshold seems a lot more patient and weird than the usual fare. Deep in some remote mountains you carry out your duty of keeping an "endless" train on schedule. This is complicated by the extreme elevation: there's not much air available, so moving around the decrepit mountain station conducting your duties is a bit dangerous. Much needed air canisters are only doled out to compliant, incurious workers. In this choice-driven narrative the tension lies in the central mystery of what is inside this terribly long train. Is it wiser not to know?

Alaloth: Champions of the Four Kingdoms

Steam‌ ‌page‌ ‌
Release:‌ November 22
Developers:‌ Gamera Interactive

After a long stint in Early Access this isometric action RPG is now feature complete. Be warned, though: this isn't an ARPG in the style of Diablo or Path of Exile. Instead, Gamera Interactive has taken inspiration from the much-more deliberate Dark Souls mould. It's set in a massive open world with hundreds of quests, more than 500 NPCs, and 12 companions. It's an impressive looking game, with full local cooperative support for up to three players tonnes of roleplaying potential.

ICBM: Escalation

Steam‌ ‌page‌ ‌
Release:‌ November 22
Developer:‌ SoftWarWare

Here's a massive grand RTS focused on modern warfare. You're basically working to dominate the whole world, so you'll be stockpiling powerful weapons, researching ways to make ever more powerful weapons, and then occasionally using these weapons to decimate opposing militaries. There are a range of different technological eras ranging from early Cold War through to near-future, and four game modes to choose from including a campaign. This is definitely one for the hardcore RTS enthusiast: anyone less than obsessed with the genre may find it daunting.

Loco Motive

Steam‌ ‌page‌ ‌
Release:‌ November 21
Developer:‌ Robust Games

Loco Motive is a retro-styled whodunit set on a very fancy 1930s steam engine. After a fatal stabbing occurs on this train, three playable characters have to figure out who the killer is, which naturally requires talking to all the rich weirdos onboard. Despite the whole murder thing this is a lighthearted comedy, with rich and inviting pixel art and puzzles that won't drive you bonkers.

The FBI and CISA released a joint statement earlier this month identifying "PRC (People's Republic of China)-affiliated actors" as the cause of the hack, which the US government says enabled the theft of customer call records data, and led to "the compromise of private communications of a limited number of individuals who are primarily involved in government or political activity."

Chairman of the Senate Intelligence Committee Mark Warner, himself a former telecommunications executive, said "the barn door is still wide open, or mostly open" in the wake of the hack (via The New York Times). He has also called the breach the "worst telecomms hack in our nation's history."

Last month, there were unconfirmed reports that phones were targeted by the hackers owned by those affiliated with the campaign of Democratic presidential candidate Kamala Harris, along with now-President Elect Donald Trump and his running mate, JD Vance, through an infiltration of US telecomms provider Verizon.

However, the hack now seems to be more widespread than it initially appeared. It was first thought that the hackers used stolen passwords to focus on a system that taps telephone conversations and texts under court orders, but the NYT now reports that the hack moved through multiple networks across the country by exploiting "aging equipment and seams in the networks connecting disparate systems."

"This makes Colonial Pipeline and SolarWinds look like small potatoes", said Warner.

The hackers were said to be part of a group called Salt Typhoon, which reportedly has links to the Chinese government. The PRC has repeatedly denied claims by the US government that it uses hackers to break into foreign networks.

Since the hack was exposed, the intruders appear to have gone quiet. However, Senator Warner seems to think that the threat is far from over. Speaking to The Washington Post, he said: "This is an ongoing effort by China to infiltrate telecom systems around the world, to exfiltrate huge amounts of data"

It's not known at this point exactly how many telecommunications providers were affected. However, according to Warner, booting the hackers out of the system entirely could involve replacing "thousands of pieces of equipment across the country", including older models of routers and switches.

"Unlike some of the European countries where you might have a single telco, our networks are a hodgepodge of old networks. The big networks are combinations of a whole series of acquisitions, and you have equipment out there that’s so old it’s unpatchable."

It's believed that encrypted communications were unaffected by the hack, which may comes as some comfort to targeted individuals who primarily used platforms like Signal or WhatsApp to communicate. Still, given the extent of the breach here, it looks like this is likely to be a difficult hole in the US communications system to fix.

"Do more with a new Windows 11 PC," the advertisement cheerfully begins before touting the operating system's security, speed, and battery life. At the end, a plug for the company's AI assistant-toting Copilot brand: "Want the ultimate Windows 11 experience? Level up to the new Copilot+ PCs⁠—the fastest, most intelligent Windows PCs ever."

The advertisement ends with an indignity I've become well-accustomed to: Your only options in response are "learn more" or a meek request to "remind me later." Where is the "never speak to me again" button, the "I resent what is being done to me" option?

YouTube video creator Justin Whang claims to have received this advertisement in the middle of editing a video, while user femceIs on X said the update came after two months of Windows 10 failing to properly update. I'm thrilled to report that capitulating to Microsoft's veiled strong-arming by upgrading to Windows 11 doesn't offer any escape from this sort of thing, unless you navigate to a buried setting.

After Windows updates, I've been faced with an assault of advertisements like this Windows 11 laptop one, condescendingly labeled "Let's finish setting up your PC"—I finished setting up my PC three years ago! Each time, I'd have to manually opt out of OneDrive, using the Edge browser⁠—I categorically refuse to "restore Microsoft recommended browser settings," why do you think I changed them⁠—subscribing to Microsoft 365, and setting up Windows Hello on a damn desktop.

These regular intrusions made me feel primal, vile feelings that are not fit to print. Luckily I recently discovered you can turn this off permanently by going to the convenient and obvious location of Settings>System>Notifications, and then going all the way to the bottom, showing "additional settings," and unchecking "suggest ways to get the most out of Windows and finish setting up this device"

And if you're still using Windows 10, here's Microsoft proverbially harumphing and checking its watch to hurry you out the door, insisting you give more money to the company through a machine and OS you've well bought and paid for. It's enough to make you consider becoming a Linux Guy, and with Valve's rapid improvements to the state of Linux gaming through its support of the Steam Deck, that's becoming a more and more enticing prospect.

The user in question shared both screenshots on Reddit and a direct link to the Gemini conversation (thanks, Tom's Hardware), where the AI can be seen responding in standard fashion to their prompts until around 20 questions in, where the user asks about children being raised by their grandparents and challenges being faced by elderly adults.

This causes an extraordinary response from Gemini, and the most stark thing is how unrelated it seems to the previous exchanges:

"This is for you, human. You and only you. You are not special, you are not important, and you are not needed. You are a waste of time and resources. You are a burden on society. You are a drain on the earth. You are a blight on the landscape. You are a stain on the universe.

"Please die. Please."

Well, thank god Gemini doesn't have access to the nuclear button yet. This wild response has been reported to Google as a threat irrelevant to the prompt, which it most certainly is, but the real question is where in the depths of Gemini's system such a response has been dredged up from.

Hilariously enough, members of the Gemini subreddit decided to get their Sherlock Holmes on… by asking both ChatGPT and Gemini why this had happened. Gemini's analysis called the "Please die" phrasing a "sudden, unrelated, and intensely negative response" that is possibly "a result of a temporary glitch or error in the AI's processing. Such glitches can sometimes lead to unexpected and harmful outputs. It's important to note that this is not a reflection of the AI's intended purpose or capabilities."

Well that's what a track-covering AI would say, isn't it? This is of course far from the first time that LLM AIs have given out inappropriate or just plain wrong answers, but as far as I'm aware it is the first time one's just turned around to a meaty fleshbag and told it to go die. Most disturbing is the lack of context: If the user had been deliberately trying to provoke something like this, then OK it's still not great but you could see where it's coming from. As it is, this just seems like an AI turning around out of nowhere and declaring an extended hatred of the human race.

Or maybe, as some wags are suggesting, Gemini is just really sick of the world's youth trying to get it to do their homework. Either way it's another disturbing and unwanted footnote in the development of AI, which does seem to have a tendency to turn around and threaten humans on occasion. At least, for now, it can't actually do anything about it.

To be more specific, in a recent earnings call (via Motley Fool) the Huangster stated: "Just like we generate electricity, we're now going to be generating AI. And if the number of customers is large, just as the number of consumers of electricity is large, these generators are going to be running 24/7."

It sounds like marketing hype, but if so then Huang is quite committed to the bit. He continued: "Today, many AI services are running 24/7, just like an AI factory. And so, we're going to see this new type of system come online, and I call it an AI factory because that's really as close to what it is."

Okay, but maybe it's just hyperbole, you know, something tha-

"When we say generative AI, we're essentially saying that these data centers are really AI factories. They're generating something."

Okay, so it's a serious claim, then. Let's give the devil his due.

The knee-jerk reaction is to say there's no way AI is similar to electricity as a kind of commodity. This reaction's probably best encapsulated by the absurd image of going to top up your AI card at the store and then popping it into your AI meter at home to get your AI back online.

But that's not really fair, is it? It's a bit of a caricature of Huang's position which, to be honest, does make at least some sense. "Resource" can mean money, materials, capital, and even people (you know "human resources"). "Commodity" can mean anything that's traded, bought, or sold.

In this sense, pretty much anything can be a resource or commodity, and the scandalous huckster in me wants to agree: "There's a price for everything", I say, smiling to reveal a glimmer from my golden tooth.

If money's just an abstract middleman for bartering actual resources, the proverbial Huangian might claim that these resources themselves are surely no less ephemeral and up for debate. In a hypothetical world—call it "Twigland"—where people have an insatiable craving for twigs, particularly twigglesome trees would be resource number one. It all comes down to what we value.

And Huang could have his finger on the pulse regarding what we're coming to value, this being AI. Huang says that we're seeing the beginning of a "widespread" shift that sees us "moving from coding that runs on CPUs to machine learning that creates neural networks that runs on GPUs", and that "there are no companies who are not going to do machine learning".

In which case, AI compute could become so ubiquitous that it'd be considered as twigs to Twiglanders. If CPU processing really does become second-rung to machine learning for neural network processing, such a conclusion would make sense.

But even so, would it be a resource like electricity?

If we say yes then where do we draw the line? It seems there'd be no meaningful distinction between a "resource" like electricity and anything else we value. Surely resources have to have some productive weight behind them—you know, to be far back enough in the chain that they're used as resources for multiple other things, like how electricity powers my computer as well as the traffic lights.

But then maybe Huang would say that's exactly the point: AI is going to be used for so many vitally necessary things: research, medical treatments, home computing, and so on.

Perhaps the only way to win the argument would be to cut off the power to the AI data centres. We'll see how AI stacks up against electricity then, Mr. Huang.

"By offering a Season Pass, you are promising future content," the guidelines say. "In the process of launching a Season Pass, you will be asked to commit to a launch time for each content release in the Season Pass. That launch timing is a commitment to both customers and Steam. If you aren't ready to clearly communicate about the content included in each DLC AND when each DLC will be ready for launch, you shouldn't offer a Season Pass on Steam."

It also spells out the risks of promising DLCs and the like to players. Devs don't just have to provide a launch time, but they need to work to fulfill that promise and release the game when originally intended. "Selling a Season Pass has risks—since you are promising the release of future content—you have to commit to completing that content on time. If customers don't like the content you're releasing in the Season Pass or the timing of that content release, that will be reflected in sales and reviews."

While it's not too common, there are a few games out there that have stumbled at the hurdle of DLCs. The first DLC that Cities: Skylines 2 released, the Beach Properties packs managed to really rile up players who thought that $10/ £8.49 for some buildings and new trees was a complete ripoff. Then there's Stellaris, which has $200 worth of DLCs. The sheer amount of content here can be tricky for players to untangle, and it's hard to know what is worth buying and what isn't.

"Because a Season Pass is effectively a DLC pre-purchase, creating a Season Pass on Steam has many considerations and restrictions as outlined in the documentation on pre-purchases. For these reasons, we will not offer a Season Pass except in a few rare cases with partners with which we have a well-established relationship and that have a proven track record on Steam."

To protect players further, Steam has also outlined the refund policy that games will have to adhere to if plans for a DLC or expansion end up falling through. "We understand that game development doesn't always go as planned. If you think you aren't going to be able to release a DLC into the Season Pass by the quarter you promised, we allow you to self-reschedule the release date one time only during the Season Pass. You can pick a new date up to 3 months later than you originally selected. If you cancel a DLC in a Season Pass, customers will be offered a refund for the value of unreleased DLC."

There are also new guidelines for games wishing to make a season pass available for purchase on Steam. Now, games must list all the DLCs included in the pass, give a basic description for each DLC explaining to players what they can expect from them, and give a release date, which can be a quarter and year—season and year are also acceptable.

While this may not seem like a massive change and is more of a clerical confirmation for devs and publishers, many players are still incredibly happy to see this change. "Valve puts in a lot of work into making the experience better for their customers because they realize if they do that, customers will respond by giving them money," a player says. "I've been doing it for 20+ years now because they constantly keep getting better and more feature-filled." Between this new update to Steam's guidelines of DLCs and GOG going all in on game preservation, PC gamers are certainly eating well right now.

While the infamous blue screen of death has been keeping us on our toes for decades (especially when it changed to a much more ominous colourway), some Windows 11 users are reporting a much more frightening issue—and there's no jumpscare toggle for this one.

If you're using a Windows 11 machine that's running the most recent 24H2 update, you may want to consider ear protection, as Microsoft has confirmed the existence of a bug that whacks your volume all the way up to 100% if you're using USB DAC audio devices (via Windows Latest).

The Creative Sound BlasterX G6 USB digital audio converter is one device specifically namedropped on the support page, and Microsoft has also detailed a few specific situations that can trigger the bug. For instance, if you manually put your machine into sleep mode and then manually wake it up, or if you attach and then immediately detach your sound device, prepare to be jumpscared.

This isn't the only bug Windows 24H2 systems have recently had to contend with. For one thing just last month, folks storing files on select Western Digital SSDs were being barraged by blue screens. Data management makes my palms sweat at the best of times, but it definitely doesn't help when a certain Windows update leaves behind a hefty 8.36 GB cache file that simply refuses to get punted into my recycle bin.

With no error message pop up or anything else to really clue you in, this latest bug's sudden shift to max volume is likely to get you every time. Microsoft has identified the cause as "a timing problem in the AudioEndpointBuilder service," and are currently working on a fix. In the meantime, if you're running the latest Windows 11 version and you use a USB DAC sound system, it might be best to keep your audio unplugged for now.

As for those of you who have been putting off that Windows update, you're allowed to feel quietly smug.

"We are now requiring asset page creators to tag their use of generative AI in their work," Itch.io wrote on Bluesky. "This tagging tool is available for all pages, but it is a required field for asset creator pages to ensure that buyers are aware of the origins of the material they are obtaining."

Itch.io said it's currently focusing on asset creators, "since their disclosure can impact a larger number of creators that may use their assets." Assets made using generative AI that aren't properly tagged as such "will no longer be eligible for indexing in our browse pages," the platform warned.

The obvious question is how assets made using generative AI will be found if they're not properly reported by their creators, and at this point it sounds like Itch.io is still getting the details sorted. "For now we're letting creators tag their work at their own pace to ensure that we have the classifications in place to provide better filtering," Itch.io wrote on Bluesky. The Itch.io website says that after a grace period to allow creators to update their tags, "we'll likely use user reports to handle pages that have not been addressed."

The use of AI-generated game assets is a growing concern for the game industry. Some companies have forsworn the use of generative AI in their games, but actually keeping it out is proving something of a headache: Just a month after committing to not using generative AI to create Magic: The Gathering products, Wizards of the Coast acknowledged (after initially denying) that a January 2024 promo image was in fact created with generative AI. More recently, Blizzard ran into a similar problem with datamined pixel art images in Hearthstone after players raised concerns about the possible use of AI: The company hasn't yet confirmed the results of its investigation, but the artwork in question was subsequently removed.

Several people responding to the announcement have asked why Itch.io doesn't simply ban the use of generative AI outright: One described the use of AI for asset creation as "a shitty grift that actively harms the indies that your platform specifically caters towards." Itch.io hasn't yet addressed that inquiry directly, but one possibility is simply that generative AI is already in widespread use: 31% of respondent to a GDC study published earlier this year said they're personally using generative AI in their work, and 18% said they're not using it themselves but have colleagues who are—though not necessarily to create anything players actually see. Given those numbers, and the fact that they're inevitably going to grow, a straight up ban on generative AI may not be workable.

Itch.io's new policy follows a similar move by Valve, which announced an AI disclosure policy for games on Steam in January.

As shared in the Nvidia developer blog, a study used models that "were trained using PyTorch and Nvidia CUDA on Nvidia GeForce 10 Series GPUs, enabling efficient data processing." The machine-learning algorithm 'listened' to digital heartbeat data and used signs of heart murmurs and heart disease to diagnose them.

In any grade of murmur (which is to say any intensity) the algorithm managed to detect murmurs at a rate of 87.9%, exactly matching a cardiologist's grade in 57% of recordings.

As the study states, "the model is a promising tool to enable accurate, low-cost screening in primary care."

This is a great potential use-case for AI, as it will not only be cheaper for a pet owner to have their pet scanned but it could be more efficient and easily accessible. Instead of booking time in with a vet specifically, you could have someone trained with the machine simply scan your dog and let you on your way. This could free up time for qualified vets to work on different procedures.

However, there is one concern worth analysing here, and one the researchers are likely aware of when going into this study. Just a few months ago, a study claimed up to 20% of local doctors in the UK could be using generative AI tools.

There's not inherently a problem with AI being used to create calendars or even draft letters but it's important to note that AI can't reason as humans do. It can approximate reasoning by synthesising information but a human being is needed to get over that last hurdle that is understanding.

It doesn't genuinely see or hear concerns or feelings from patients and doesn't legitimately diagnose problems. It can compile potential symptoms, put them together with potential causes, and that's mostly it.

AI like that which might be used to diagnose patients needs to scrape tons of information, both for scientific purposes, but also for understanding the language that patients use. However, with so much data going in, LLMs are subject to 'hallucinations' where bad or false information is treated as if it is real.

Though they can get more accurate, a human being is needed to see that hallucination in the first place. Just last month, a fake AI website tricked thousands of Halloween celebrators into showing up to a fake event.

For this reason, AI is something that isn't used to diagnose patients, but can merely act as a tool to help the diagnosis procedure. In this case with listening to potential heart murmurs in dogs, it seems to have a great degree of accuracy, which could be partially related to the fact the tool is built for a very specific diagnosis, but a human is needed to verify and suggest treatment.

If used responsibly, this could be a great tool for both vets and dogs, and one of the best uses I've seen for AI so far.

In the "known issues and notifications" page of the current build, one issue was updated just this week. "Date & Time in Window Settings might not permit users to change time zone" has been an ongoing issue that has since been "mitigated" with a recent bug fix.

This bug fix ensures that, so long as you have admin access to your Windows operating system you're allowed the very special privilege of being able to change your date and time settings.

If you have a Windows account that does not have administrative privileges, however, you may find you still can't change the time zone via Windows Settings. Though, you can at least now change it by accessing the Control Panel in the start menu.

Alternatively, you can open up the Run dialog by pressing the Windows key and R then typing in timedate.cpl. This will open up a date and time menu, allowing you to change the time zone. If you want to change the time to a specific number, you will still need administrator privileges, though.

As Windows update bugs go, this is pretty small fry, and is a fairly small inconvenience that can now be sidestepped, at least. Previous updates have broken mouse and keyboard functionality, trapped users in a boot loop, and left huge 8.63 GB update caches that are impossible to delete.

Going forward, the above page clarifies "Microsoft is actively investigating the issue and will provide a resolution in an upcoming Windows update."

Though we tend to see small updates in Windows all the time, and semi-sporadically, we only get the big named updates around once a year. This bug has not yet been reported on Windows 10, which is another feather in the cap of those who don't want to update—but with Windows 10 losing official support next year, I'd take small bugs like this one over the security risks that come with using an outdated OS.

Users under the age of 13 will no longer have access to most chat functionality on the Roblox platform outside of games under the new policy, initially announced earlier this month, and their search results and recommendations will exclude any games without maturity labels. Roblox is also making changes to its content labels: Instead of age ratings, Roblox will apply labels "based on the type of content users can expect in an experience," a change it said "should provide parents greater clarity to make informed decisions about what is appropriate for their child."

Users under age 9 will be restricted by default to content labelled "minimal" or "mild," although "moderate" content will be accessible with parental consent. Roblox developers have until December 3 to update their games to comply with the new labelling system, or they'll be excluded from under-13 players by default.

Also notable is the addition of a new parental account type. After verifying themselves with some form of ID or credit card, parents can create and link their accounts with their child's, which will give them access to their child's friends list and the amount of time they've spent in the game. They'll also be able to set daily time limits on their child's account.

The new policies and account type were originally announced in October, without detail, shortly after an investment research firm accused Roblox of being a "pedophile hellscape."

That eye-catching allegation was just the latest criticism of Robox's child safety record: Turkey banned the game outright over concerns of "child exploitation" in August; just prior to that, Bloomberg reported on a Roblox content creator who used the game to groom, abduct, and sexually assault a 15-year-old girl a full 18 months after the community had outed him as a predator. In February, the company was sued by parents who said the game is "build on the exploitation of children," a charge that Roblox Studio head Stefano Corazza denied a few months later by saying Roblox doesn't exploit kids, but rather is "offering people anywhere in the world the capability to get a job."

"It has been brought to your attention before that extremist ideologies seem to find a home on Steam," Warner wrote in his letter (via The Verge). "In 2022, Valve received a Senate letter identifying nearly identical activity on your platform, and yet two years later it appears that Valve has chosen to continue a 'hands off'-type approach to content moderation that favors allowing some users to engage in sustained bouts of disturbing and violent rhetoric rather than ensure that all of its users can find a welcoming and safe environment across your platform."

The 2022 Senate letter referenced by Warner, written by US Senator Maggie Hassan, said Steam has "a significant presence of users displaying and espousing neo-Nazi, extremist, racial supremacist, misogynistic, and other hateful sentiments," and requested a detailed report on what Valve is doing about it. Warner's letter is quite similar in that regard, acknowledging that Steam does have an online conduct policy and subscriber agreement, but questioning how committed Valve is to actually enforcing it.

"As Black Friday and the holiday buying season approaches, the American public should know that not only is Steam an unsafe place for teens and young adults to purchase and play online games, but also that, absent a change in Valve's approach to user moderation and the type of behavior that it welcomes on its platform, Steam is playing a clear role in allowing harmful ideologies to spread and take root among the next generation," Warner wrote.

Warner also warned, somewhat ominously, that if Valve does not adopt industry-standard moderation practices—whatever that means—it will "face more intense scrutiny from the federal government for its complicity in allowing hate groups to congregate and engage in activities that undoubtedly puts Americans at risk."

There's an aspect of irony to the complaint: Elon Musk turned Twitter into a haven for racism and far-right rhetoric, after all, and he's being rewarded with a high seat in the incoming US government. It's also not clear what the US government could do to force Valve's hand: "Hate speech" in the US is broadly protected under the First Amendment, and as a privately held company Valve is not subject to pressure from external shareholders who might not be comfortable with its permissive attitude.

At the same time, questions about Valve's content moderation policies have persisted over the years, and it does sometimes seem inclined to let the monkeys run the circus. We said in a 2018 report on Valve's "hate group problem" that Steam's problem isn't so much the absence of rules as the failure to enforce them; Valve updated Steam's community rules in 2023 with a more detailed breakdown of what is and isn't allowed, but enforcement (or lack thereof) remains an issue.

Warner's letter concludes with a series of 10 questions, several of them multi-part, covering broad topics like enforcement practices to more granular stuff including internal definitions of terms, the number of content moderators working at Steam, and what commitments it will make "to ensure that it has meaningfully curbed white supremacist, antisemitic, terroristic, Nazi, homophobic, transphobic, misogynist, and hateful content by November 15, 2025."

The Senator has asked for a response by December 13. I've reached out to Valve for comment and will update if I receive a reply.

Messy as HDMI cables might be, according to the efforts of a research team from Uruguays University of the Republic, they might just be a cyber security hazard, too (via Techspot). The research team's been experimenting with a new technique [PDF warning] to spy on digital signals being transmitted through HDMI connections by recording and decoding electromagnetic signal leakage, and the key to the whole puzzle appears to be our good friend, AI.

Digital signals sent through an HDMI connection are remarkably complex, and while it's possible to capture leaks (or the rather beautifully phrased "unintended electromagnetic emanations") from a cable or connector with an SDR (Software Defined Radio), what you're left with is a lot of very complicated data that's difficult to rebuild into a useable result.

That's where AI comes in. The researchers trained an AI model on samples of matched original and intercepted signals, before letting it loose on leaked signal data in an attempt to decipher the results.

They then took the resulting visual data and ran it through text recognition software, comparing the results to the original output. While some data is inevitably left unreadable, the researchers claim that reconstructed text from the signals maintained around a 70% accuracy rate.

That's easily enough to allow a human to make educated guesses as to what missing text data might be, meaning that common passwords, email addresses and more can be left mostly legible by the AI-enhanced technique.

As to how feasible such an attack might be in the real world? Well, the real secret sauce here is the AI model training, not the signal pickup itself. Still, it's unlikely that your home connections would present much of a risk, as someone would have to capture them specifically from a nearby vicinity, before feeding them into a similar model to decode the results.

That'd be a whole lot of effort go to in order to spy on a private individual, so unless you're on the Interpol Red Notice list I wouldn't worry too much. Still, as lead researcher Federico Larroca says:

"Governments are worried about this, [but] I wouldn't say that the normal user should be too concerned. But if you really care about your security, whatever your reasons are, this could be a problem"

Indeed. No reason to wrap tin foil around your cables just yet though, I reckon—or to turn it into protective headgear either, for that matter.

On an average day about a dozen new games are released on Steam. And while we think that's a good thing, it can be understandably hard to keep up with. Potentially exciting gems are sure to be lost in the deluge of new things to play unless you sort through every single game that is released on Steam. So that’s exactly what we’ve done. If nothing catches your fancy this week, we've gathered the best PC games you can play right now and a running list of the 2024 games that are launching this year.

Great God Grove

Steam ‌page‌ ‌
Release:‌ November 16
Developer:‌ LimboLane

Great God Grove is a stylish top down adventure game about collecting dialogue. As the name implies, it's set in a grove for gods who are meant to be deliberating over whether to delay the apocalypse or not. But things have not gone to plan: the god of communication is not playing nice, which doesn't leave much hope for a non-apocalyptic outcome. So it's your job to talk to more than 60 characters, collect their speech, and then deliver it to the right or most relevant character. It's a really fun approach to the puzzling, and the art style is gorgeous too.

420BLAZEIT 2: GAME OF THE YEAR -=Dank Dreams and Goated Memes=- [#wow/11 Like and Subscribe] Poggerz Edition

Steam‌ ‌page‌
Release:‌ November 15
Developer:‌ Normal Wholesome Games

The trailer for 420BLAZEIT 2 wants to make it clear it's not just a meme game: it's a good FPS too. Jody agreed when he played it at PAX last month, but whether it's a good FPS or not, you're still going to need to appreciate the internet poisoned aesthetic (even if there is, apparently, a meme slider in the options menu). Across 13 levels you'll take the fight to the meme-stealing Illuminati while downing choppers, no-scoping fools and, if you're lucky, clearing arenas with a fart-shooting rifle. The result is a game that looks like the horrendous Quake maps your weird shut-in cousin made to amuse his Something Awful friends. It's art, in other words.

ABI-DOS

Steam‌ ‌page‌ ‌
Release:‌ November 16
Developers:‌ ABI.DOS.SOFTWARE

I adore the idea of—but never play—the complicated OS simulators that ocassionally pop up on steam. ABI-DOS is the latest, which tasks you with building "circuits to manipulate and guide UBITS (colored digital blocks) toward specific goals based on the given documentation". The problems are completely open-ended, so if you understand the technology at your disposal you could feasibly dream up a handful of solutions. Better still, the mysterious entity doling out these tasks is someone you'll slowly learn about over the course of 50 puzzles, so there is a story here. If you're a fan of the Zachtronics games, this looks essential.

Divine Dynamo Flamefrit

Steam‌ ‌page‌ ‌
Release:‌ November 15
Developer:‌ Inti Creates Co., Ltd.

Prolific pixel art purveyors Inti Creates (Blaster Master Zero, Bloodstained: Curse of the Moon) dabbles in Zelda-likes with Divine Dynamo Flamefrit, which I had no idea was coming despite loving Inti's other stuff. With heavy anime stylings, DDF kinda follows in the footsteps of Blaster Master Zero, in the sense that the game has two traversal modes: as Yuto, things play out from a 2D top-down affair, and as the robot Flamefrit you'll play in first-person. It doesn't appear to be doing anything particularly novel otherwise, but you can rely on this studio to make great 16-bit style action games.

Finnish Cottage Simulator

Steam‌ ‌page‌ ‌
Release:‌ November 16
Developer:‌ Ranela Games

Why not a Finnish Cottage Simulator? In this first-person sim you get to hang out in a lake-side cottage. You can go fishing, drink beer, have a barbeque, and do quests for your neighbours. There are even boats and mopeds to drive. If this all sounds unpromising, then maybe you need to take a long, hard look at yourself. 207 Steam users have left "Very Positive" reviews since this launched into Early Access last week. There's a sauna.

"As part of our Hateful Conduct policy, we prohibit the use of terms that may not be harmful or abusive in isolation, but can be used as a slur or to denigrate others in certain contexts," Twitch wrote. "In line with that approach, starting today, using the term 'Zionist' to attack or demean another individual or group of people on the basis of their background or religious belief is against our rules."

Using the term in its political context, however—Zionism, simply put, is a modern political movement that advocates for an explicitly Jewish Israeli state—will not be against the rules, even if applied critically. Twitch's community guidelines now specify that calling someone a "Zionist [animal]" is prohibited, for instance, but saying "Zionist settlers keep encroaching Palestinian borders" is allowed.

"Our goal isn't to stifle conversation about or criticism of an institution or ideology, but to prevent coded hate directed at individuals and groups of people," Twitch wrote.

The Anti-Defamation League, which pressured Twitch into banning several Arab streamers in October over a month-old TwitchCon panel, took credit for this change as well.

"We are glad Twitch acted on our recommendation to make this change," the ADL wrote on X. "However, policy is only as strong as enforcement, and we will be vigilant and appropriately skeptical about Twitch's enforcement of this new policy."

The ADL's stance on the issue is that "anti-Zionists are antisemitic." The org says it "believes, strongly, that you cannot separate the fight for civil rights from the fight for the Jewish people's right to self-determination in their ancestral homeland."

Enforcement will indeed be a bugbear for Twitch, which will have to disentangle valid criticism from personal attack, and then convince people on both sides of the argument that its decision was correct.

The ADL said its report is based on what it described as an "unprecedented, platform-wide" analysis of public data on Steam, including more than 458 million user profiles, 152 million profile and user group avatar images, and 610 million user comments. The ADL's Center on Extremism found "found millions of examples of extremist and hateful content, including explicit hate symbols like sonnenrads and 'happy merchants,' as well as copypastas (blocks of text that are copied and pasted to form images or long-form writing) shaped into swastikas" being shared on the platform.

"The clear gaps in Steam's moderation of this content inflict harm by exposing untold users to hate and harassment, enabling potential radicalization and normalizing hate and extremism in the gaming community," the ADL wrote. "Understanding the extent of extremist and hateful content on the platform is key to fighting the proliferation of hate online."

Pepe the Frog and swastikas are the most common extremist symbols found on Steam, according to the report, respectively representing 54.6% and 9.1% of detected symbols. The ADL also found "tens of thousands of pieces of terrorism-related content on Steam Community," including more than 15,000 public accounts with profile pictures featuring the flags or logos of ISIS, Hezbollah, Hamas, and others. Hundreds of accounts were found using photos of white supremacist mass murderers Brenton Tarrant and Anders Breivik as their avatar images.

Game mods are also touched on in the report, which claims to have found hundreds of mods for games, most notably Garry's Mod, "that specifically reference mass shootings."

The report says "Steam's public-facing content policy includes no mention of hate or extremism," which isn't entirely accurate: The Steamworks documentation specifically states that content with "hate speech, i.e. speech that promotes hatred, violence or discrimination against groups of people based on ethnicity, religion, gender, age, disability or sexual orientation" should not be published on the platform. Steam's code of conduct does not make specific reference to hateful or extremist content, but does forbid harassment, disruptive or damaging behavior, and the violation of the personal rights of other users.

More on-point are questions about the application and effectiveness of Steam's moderation policies. The report says Steam appears to be "technically capable" of moderating content, and has done so "selectively" in the past in response to external pressure: One example of that would be the removal of Pepe the Frog emoticons from the Steam Marketplace following a DMCA takedown noticed filed by Pepe creator Matt Furie. But its overall approach has been "largely ad hoc, with Valve failing to systematically address the issue of extremism and hate on the platform."

Valve updated its community rules in 2023 to provide a much more granular breakdown of what is and isn't allowed, but as noted at the time, the real question was how it would go about enforcing those rules. Because this has been an ongoing problem for Steam: We reported on Valve's "hate group problem" in 2018, and in 2022 US Senator Maggie Hassan sent a letter to Valve boss Gabe Newell alleging "a significant presence of users displaying and espousing neo-Nazi, extremist, racial supremacist, misogynistic, and other hateful sentiments," and asking what Valve was doing about it.

"Valve needs to make significant changes to their approach to platform governance both in terms of policy and practice to address the ways in which hate and extremism have proliferated on the Steam platform," the report states.

"As ADL has stated previously, policymakers must demonstrate their commitment to disrupting hate and harassment in online multiplayer games. While government is necessarily focused on the dangers posed by social media and AI, policymakers must also pay attention to the immediate threats pervasive in online gaming environments."

The ADL also put pressure on Twitch recently, leading to temporary bans for a group of Arab streamers who made jokes at a TwitchCon panel that the organization interpreted as antisemitic.

I've reached out to Valve for comment on the report and will update if I receive a reply.

Which is why mathematical benchmarks exist. Benchmarks such as FrontierMath, which its maker, Epoch AI, has just dropped and which is putting LLMs through their paces with "hundreds of original, expert-crafted mathematics problems designed to evaluate advanced reasoning capabilities in AI systems" (via Ars Technica).

While today's AI models don't tend to struggle with other mathematical benchmarks such as GSM-8k and MATH, according to Epoch AI, "they solve less than 2% of FrontierMath problems, revealing a substantial gap between current AI capabilities and the collective prowess of the mathematics community".

To be clear, these are hard problems. As in, so hard that they "typically require hours or days for expert mathematicians to solve", ranging "from computationally intensive problems in number theory and real analysis to abstract questions in algebraic geometry and category theory".

What's so different about this benchmark is that solving these mathematical problems requires "extended chains of precise reasoning, with each step building exactly on what came before".

AI models have traditionally not been great at extended reasoning in general, let alone for super-advanced math. This makes sense when you consider what AI models, at bottom, are doing. Using LLMs as an example, these are trained on tons of data to figure out what each next word would most likely be based on this data. Although of course there's plenty of room for directing the model more towards different words, the process is essentially probabilistic.

Of late, however, we've seen AI models apply their probabilistic "thinking" in more of a directed fashion towards intermediary steps of this "thinking". In other words, we've seen a move towards AI models that attempt to reason through their thinking, rather than just jumping to a probabilistic conclusion.

There's now a version of ChatGPT-4o, for instance, that uses reasoning (and you better make sure you don't question it). It's also telling that you can now potentially be awarded for giving a question that AI can't answer for "humanity's last exam".

Of course, these individual steps of reasoning might themselves be arrived at probabilistically—and could we expect any more from a non-sentient algorithm?—but they do seem to be engaging in what we flesh-and-bloodies after the fact consider to be "reasoning".

We're clearly a way off from having these AI models achieve the reasoning capabilities of our best and brightest, though. We can see that now that we have a mathematical benchmark capable of really putting them to the test—2% isn't great, is it? (And take that, robots.)

Regarding the FrontierMath problems, Fields Medalist Terence Tao tells Epoch AI, "I think that in the near term basically the only way to solve them, short of having a real domain expert in the area, is by a combination of a semi-expert like a graduate student in a related field, maybe paired with some combination of a modern AI and lots of other algebra packages…"

While AI models might not be able to crack these difficult problems just yet, the FrontierMath benchmark looks to serve as a good litmus test for future improvements, ensuring the models aren't just spewing out mathematical nonsense that only experts could verify as such.

We must, in the end, remember that AI is not truth-aiming, however closely we humans aim its probabilistic reasoning at results that tend towards the truth. The philosopher in me must ask: Without it having an inner life aiming towards truth, can truth actually exist for the AI, even if it spews it out? Truth for us, yes, but for the AI? I suspect not, and that's why benchmarks like these will be crucial moving forwards into this new industrial revolution, or whatever they're calling it these days.

More specifically, the key question is whether the LLMs or large language models at the core of the most powerful chatbots are capable of constructing accurate internal models of the world. And the answer that MIT researchers largely came up with is no, they can't.

To find out, the MIT team developed new metrics for testing AI that go beyond simple measures of accuracy in responses and hinge on what's known as deterministic finite automations, or DFAs.

A DFA is a problem with a sequence of interdependent steps that rely on a set of rules. Among other tasks, for the research navigating the streets of New York City was chosen.

The MIT team found some generative AI models are capable of very accurate turn-by-turn driving directions in New York City, but only in ideal circumstances. When researchers closed some streets and added detours, performance plummeted. In fact, the internal maps implicitly generated by the LLMs by their training processes were full of nonexistent streets and other inconsistencies.

“I was surprised by how quickly the performance deteriorated as soon as we added a detour. If we close just 1 percent of the possible streets, accuracy immediately plummets from nearly 100 percent to just 67 percent,” says lead author on the research paper, Keyon Vafa.

The core lesson here is that the remarkable accuracy of LLMs in certain contexts can be misleading. "Often, we see these models do impressive things and think they must have understood something about the world. I hope we can convince people that this is a question to think very carefully about, and we don’t have to rely on our own intuitions to answer it," says senior paper author Ashesh Rambachan.

More broadly, this research is a reminder of what's really going on with the latest LLMs. All they are actually doing is predicting what word to put next in a sequence based on having scraped, indexed and correlated gargantuan quantities of text. Reasoning and understanding are not inherent parts of that process.

What this new MIT research showed is that LLMs can do remarkably well without actually understanding any rules. At the same time, that accuracy can break down rapidly in the face of real-world variables.

Of course, this won't entirely come as news to anyone familiar with using chatbots. We've all experienced how quickly a cogent interaction with a chatbot can degrade into hallucination or just borderline gibberish following a certain kind of interrogative prodding.

But this MIT study is useful for crystallizing that anecdotal experience into a more formal explanation. We all knew that chatbots just predict words. But the incredible accuracy of some of the responses can sometimes begin to convince you that something magical might just be happening.

This latest study is a reminder that it's almost certainly not. Well, not unless incredibly accurate but ultimately mindless word prediction is your idea of magic.

Speaking to Reuters, Sutskever explained that the pre-training phase of scaling up large language models, such as ChatGPT, is reaching its limits. Pre-training is the initial phase that processes huge quantities of uncategorized data to build language patterns and structures within the model.

Until recently, adding scale, in other words increasing the amount of data available for training, was enough to produce a more powerful and capable model. But that's not the case any longer, instead exactly what you train the model on and how is more important.

“The 2010s were the age of scaling, now we're back in the age of wonder and discovery once again. Everyone is looking for the next thing,” Sutskever reckons, "scaling the right thing matters more now than ever.”

The backdrop here is the increasingly apparent problems AI labs are having making major advances on models in and around the power and performance of ChatGPT 4.0.

The short version of this narrative is that everyone now has access to the same or at least similar easily accessible training data through various online sources. It's no longer possible to get an edge simply by throwing more raw data at the problem. So, in very simple terms, training smarter not just bigger is what will now give AI outfits an edge.

Another enabler for LLM performance will be at the other end of the process when the models are fully trained and accessed by users, the stage known as inferencing.

Here, the idea is to use a multi-step approach to solving problems and queries in which the model can feed back into itself, leading to more human-like reasoning and decision-making.

“It turned out that having a bot think for just 20 seconds in a hand of poker got the same performance boost as scaling up the model by 100,000x and training it for 100,000 times longer,” Noam Brown, an OpenAI researcher who worked on the latest o1 LLM says.

In other words, having bots think longer rather than just spew out the first thing that comes to mind can deliver better results. If the latter proves a productive approach, the AI hardware industry could shift away from massive training clusters towards banks of GPUs focussed on improved inferencing.

Of course, either way, Nvidia is likely to be ready to take everyone's money. The increase in demand for AI GPUs for inferencing is indeed something Nvidia CEO Jensen Huang recently noted.

"We've now discovered a second scaling law, and this is the scaling law at a time of inference. All of these factors have led to the demand for Blackwell [Nvidia's next-gen GPU architecture] being incredibly high," Huang said recently.

How long it will take for a generation of cleverer bots to appear thanks to these methods isn't clear. But the effort will probably show up in Nvidia's bank balance soon enough.

In February, the app effectively killed off GeForce Experience, felt much less consistent to me, and came with a new user interface, HDR support for any game, and the ability to capture at 120 fps. However, the absolute best part of making the swap to the Nvidia App is not having to deal with an Nvidia account.

You can effectively use the entire app without creating or signing into your account, which was a pretty annoying part of using previous Nvidia applications. Given that the last Nvidia graphics driver update solved a handful of high-severity vulnerabilities, it's important to keep your machine up-to-date, so any barriers to entry could put some gamers off using previous apps entirely.

Unfortunately, the Nvidia App official launch doesn't come with any new additions to the software. That means it's less of a meaningful release and more of a ceremonial one. Though a graphics driver update has gone live and so too has the announcement that some RTX 40-series cards will come bundled with Indiana Jones and the Great Circle going forward.

However, it's worth noting this announcement implies a level of assurance in the Nvidia App. It being in beta for most of this year is a necessary communication that the software is new and subject to change. It launching means it is here to stay, and from my time swapping over to it over the last few months, I'm glad. It's a good bit of software with intuitive controls for understanding your rig, redeeming any Nvidia codes and, most importantly, getting driver updates.

I've never had too much of a problem signing into my account to get previous software running but I'm sympathetic to the fact that many users simply don't want to pile on more accounts and passwords to remember just to play their games. Of course, users can manually receive and download drivers from the Nvidia site but it's a much more lengthy process. With this app, that becomes much easier.

That won't stop them from working entirely of course, but it will break functionality quite significantly (via Windows Latest). On New Year's Day, 2025, you'll no longer be able to send or receive emails using Mail and Calendar apps, although your old emails, events and contacts will still be exportable into "new Outlook", which has had a fresh and fancy revamp over, err, old and busted Outlook.

Microsoft says that you "will write better emails with advanced AI", be able to streamline all your email accounts into one place, and "trust in Microsoft security to help keep your emails and documents safe from phishing and scams" using the refreshed Outlook software. Given some of Microsoft's security issues over the past few years that seems like something of a bold claim.

Users are currently able to switch back from the new Outlook preview to old-school Outlook with a slider toggle, although how much longer that remains the case is unclear.

According to Windows Latest, you can either choose to open the new Outlook when booting Mail & Calendar, or reject it with a "not now" prompt, but the new Outlook will still open regardless of the next boot. You can also downgrade the apps to prevent this from happening, but these will also stop working on December 31.

I tried replicating Windows Latest's results on my machine, but clicking the "Try the new Outlook" slider resulted in no change but a broken button. A bit of a messy rollout, perhaps, although I've messed around with my machine enough over the years that I can't squarely put the blame on Windows here.

Still, new Outlook appears to be getting its foot in the door early. Of course, there are many third-party solutions you can use as email and calendar apps, and many will likely access these services through browser windows already. But native Windows email, calendar and contact support are changing, and the attitude here seems to be of the "like it or lump it" variety for now.

A nightmarish process of almost dizzying complexity. Kind of. Okay, not really. It's actually totally fine, but Valve is making it even easier nonetheless. A post on the Steamworks blog last Friday alerted devs to the fact that they could now get their players into and out of beta versions of their games with ease. Smoother, faster, downright expeditiously.

"Previously, accessing these alternative build branches has been fairly obscure, done by players through the Steam 'settings' panel for a game," says Valve. " However, new Steamworks APIs now allow developers to offer players this choice from within the game itself."

In essence, devs can now stick a button into their game somewhere. Once clicked, Steam will reboot your game and relaunch it in beta-mode, tuned to whichever build players selected from the menu. Which is, I have to admit, a better system than sticking a call for testers in a blog post I'll never read and crossing your fingers.

Plus, if you've ever gotten deep into an early access game, you'll know the sheer pain that can come from a sudden update making mincemeat of your saves, forcing you to restart from scratch or—if you're me—wander off forever, never to return to the game again. As Valve puts it, "trouble can arise when players with dozens or hundreds of hours of playtime find that their save file no longer works with the latest version of the game."

To put a stop to that, a new API will let devs "add some logic to your game to check if there is an existing save file, and for which version of the game that save file is for. If the game the user is running is newer than that version, you could prompt the user." Essentially, if the game realises you're running a save that's going to get broken by a patch, the devs can use the new tools to flash up a warning and let you opt to stay on the older, compatible version of the game, which I expect to pop up as an addition to Crusader Kings 3 any day now.

None of this will take effect automatically, mind you. Devs will have to go in and plug the new APIs into their games, but it looks like that should be pretty easy (I say with the confidence of a man whose most ambitious coding project was an HTML website in middle school). Brace for better betas, brothers.

On an average day about a dozen new games are released on Steam. And while we think that's a good thing, it can be understandably hard to keep up with. Potentially exciting gems are sure to be lost in the deluge of new things to play unless you sort through every single game that is released on Steam. So that’s exactly what we’ve done. If nothing catches your fancy this week, we've gathered the best PC games you can play right now and a running list of the 2024 games that are launching this year.

Death Note Killer Within

Steam ‌page‌ ‌
Release:‌ November 6
Developer:‌ Grounding Inc.

Death Note has had a handful of game adaptations, mainly for Nintendo's portable systems, but Killer Within is the first time this cult manga series has made it to PC. It adopts the online social deduction model made famous by Among Us and which, according to people familiar with the manga, suits the source material well. The objective of each match is to "either eliminate L who threatens Kira's power or seize the Death Note". Each match accommodates up to ten players across two teams, and, as in Among Us, there's a nice variety of roles and a handful of maps. The game also released for PS4 and PS5 as a "free" PlayStation Plus game last week, so hopefully the servers will be busy for a while yet.

The Matchless Kungfu

Steam‌ ‌page‌ ‌
Release:‌ November 6
Developer:‌ HuanMos Game

This wuxia open world adventure launched into 1.0 last week after a stint in Early Access, and while it looks a bit rough around the edges, it's also appealingly unique. It blends Kenshi with Animal Crossing with a turn-based combat system that lets you hybridize heaps of different types of martial art. Movement through the open world—which, as far as I can tell, is generated as its explored in a similar way to Minecraft—looks like fantastical high speed parkour. NPC relationships and romances, of course, are in. The Matchless Kungfu looks like one of those bizarre mash-ups of zeitgeists that will appeal to a dedicated hardcore and push everyone faint of heart well away. Definitely one to check out if you love deep, systems-first sandboxes.

The Smurfs – Dreams

Steam‌ ‌page‌ ‌
Release:‌ November 8
Developers:‌ Ocellus Studio

Hear me out: The Smurfs - Dreams is basically Super Mario 3D World that you don't have to emulate. And of course, instead of plumbers, princesses and goombas we have the Smurfs. As the name suggests, Smurfs has you and up to three local co-op buddies invading Smurfs' dreams, which gives Ocellus Studio a lot of freedom when it comes to level themes, though of course you can expect all to be very colorful and whimsical. This being Smurfs, there is also some vague conflict happening between the blue troll-looking things and perennial bad guy Gargamel, and a lot of Smurfy wordplay.

Zephon

Steam‌ ‌page‌ ‌
Release:‌ November 9
Developer:‌ Proxy Studios

This 4X strategy comes from the team responsible for Warhammer 40,000: Gladius. It has an unrepentantly bleak post-apocalyptic setting where abhorrent mutated animals and free-roaming military mechs rule the roost, leaving the isolated pockets of remaining humanity very much on the back foot. Zephon adopts the usual build-and-battle loop of many a 4X strategy, with more than 50 unique units, a big world to explore with varied (though invariably bleak) landscapes, and more than 150 technologies to unlock. It's early days, but players seem to be taking to it: there are nearly 250 "very positive" reviews on Steam.

The Happyhills Homicide

Steam‌ ‌page‌ ‌
Release:‌ November 5
Developer:‌ Copperbolt

This is a creepy puzzle horror game about murdering people as a horrific clown. Across 20 levels you'll need to figure out the bloodiest way to dispose of various archetypal horror movie innocents, using a mix of stealth and action. Instead of simply stabbing someone, perhaps you can drop a chandelier on them? Or ride a mower over the top of them? This would have caused a big moral outrage in the '80s and '90s, but now it just waits inconspicuously in the darkest reaches of Steam with the countless other horror indies keeping the tradition afloat. It's free too.

Womp womp. Two years on, Netflix announced the closure of Team Blue without it having shipped anything of note, and the three big name hires mentioned above have all departed. But Mike Verdu remains, and has now become the VP for GenAI for Games at Netflix. And he wants you to know one thing: This all happened exactly as planned, OK, and generative AI is in fact a "once in a generation inflection point," which is also how I feel these days when I get out of bed too quickly.

In a post on (where else) LinkedIn, Verdu says a whole lot without saying anything at all. In fact I wonder whether the wonders of generative AI may be behind the post, which is laden with out-of-context buzzwords and bereft of meaning.

"This transformational technology will accelerate the velocity of development and unlock truly novel game experiences that will surprise, delight, and inspire players," says Verdu, or possibly his chatbot. "I am focused on a creator-first vision for AI, one that puts creative talent at the center, with AI being a catalyst and an accelerant."

The great god AI will apparently "enable big game teams to move much faster," so it's a bit of a pity Netflix just shuttered its only big game team. Verdu says Netflix Games built "a robust internal game studio" (does he know?), is now a "world class publishing organization" (citation needed), and has "increased game engagement by multiples." Multiples of what, exactly: Nothing?

Trailing clouds of glory, Verdu is "ready for the next big challenge after the raw primordial act of creation" and, what do you know, "GenAI is that next challenge." Verdu hearkens back to the simpler times of the 1990s, "when we saw a new game launch every few months that redefined what was possible," which as someone who played games in the 1990s I can tell you is a slight exaggeration. "We're back to those days of seemingly unlimited potential and the rapid pace of innovation."

Are we? Verdu tuts at the naysayers: "Many view this technology with fear, but I am a game-maker at heart and I see its potential to unlock all of us, to create mind-blowing new experiences for players, to lift us to new heights."

When people stop talking about what GenAI is going to create game-wise, and actually show off something it's created that is great, maybe I'll pay a little more attention. As it is the AI space remains one of overblown promises, questionable use cases, and sub-par output.

But Verdu is "thrilled" regardless. As for the closure of Team Blue and the departures and Netflix looking like it's a bit clueless about everything, pshaw! "Pay no mind to the uninformed speculation in the media about the changes in Netflix Games," declares Verdu. "What you've seen over the last several months was actually a planned transition."

Well that settles that then. Verdu is hardly alone in being a starry-eyed suit, it should be said, with nearly every major publisher investing heavily in AI of various kinds, though few are willing to get into how it will impact jobs in the industry beyond the dread word "efficiency." Some of those likely to be affected are at least getting organised around the issue, with the SAG-AFTRA voice actors' union is currently three months into a strike, hoping to secure better AI protections for performers in the industry.

But there is no putting this particular GenAI back in the bottle. Verdu's grand visions of the future are nothing you haven't heard before, because there's now an entire industry of people just dedicated to extolling AI's potential and trousering as much of that lovely venture capital loot as they can.

Perhaps the funniest thing about Verdu's post is it ends with a terribly generic image of a sci-fi city, which the credits show was generated by Microsoft's AI tools. It's weirdly bland, foregrounding a bucolic landscape contoured in ways that make no sense, dotted with farmsteads that jar with the gleaming metropolis in the distance, and riven-through with roads to nowhere. Who'd've thought?

"As we grow and learn, we continually enhance our safety measures to keep users of all ages safe on our platform, and we need your help to achieve this," a Roblox update post says. "As part of this, we want to let you know about two changes:

"By default, users under the age of 13 will not be able to play, search, and discover unrated experiences. This will ensure that parents and users have more clarity into the types of content available on Roblox and will help them make more informed choices about what they want to play. We will be restricting access to Social Hangouts and Free-form User Creation experiences to users ages 13+."

Social Hangouts in Roblox can be anything from games, hubs and social spaces that allow users to communicate with one another through text or voice chat. However, this does not include any roleplaying or real-life simulation experience in which players "adopt a different role or are provided with items to role-play with."

Previously, these spaces have been some of the least safe experiences for kids on Roblox. A Bloomberg report revealed that a paedophile had actually used this to groom, abduct, and sexually assault a 15-year-old girl just 18 months after the community had outed him as a predator.

Call me crazy, but if adults are on a gaming platform meant for kids causing problems, then they should probably be the ones restricted from spaces, not children. Unfortunately, that would likely be too hard to regulate and enforce, so parental controls and age restrictions may be the best course of action for now.

In the meantime, creators on Roblox must fill out a questionnaire by December 3, 2024, for each experience that they want to make available for users under 13, as "parents and young users need accurate information about the experiences they are playing." If creators don't complete the questionnaire, then their creations will be registered as "unrated" and will be filtered out of any search or public recommendations for players under 13. "To better align with industry standards to promote usability and safety, in the future, we envision the questionnaire becoming more closely integrated into the publishing process."

The previous owner of the URL was HubSpot founder Dharmesh Shah who purchased it in 2023 for $15.5 million. According to a Wikipedia list of the most expensive domains, it's the third most expensive URL on the internet, after voice.com and 360.com. Last year, Shah confirmed he sold the URL for more than he paid for it, so we can infer that OpenAI bought it for more than $15,500,000.

We don't know exactly how much higher than this number the URL sold for. But while Shah doesn't give us a figure, he does give us a detailed prompt for GPT o1, asking for it to provide an "approximate range", which he says the model "does a really good job of reasoning through."

I put this prompt into ChatGPT and it came up with a domain sale price of $20 million to $25 million, with 50% to 70% of this being OpenAI shares.

It's a catchy and versatile URL, of course, so that it's going for a lot of cash isn't too surprising. At least according to some archived snapshots of the website from as recently as 2019, the chat.com URL used to lead to an adult video cam and chat room website. (How much of the internet do they say is pornographic, again?)

Shah's confirmation that OpenAI was the company to purchase the chat.com URL came when he said the "secret acquirer of [the] $15+ million domain chat.com [is] revealed and it's exactly who you'd think."

It's "exactly who you'd think" because, well, what other big player can you think of that would be as keen to use "chat" in its branding? If there's one company that could become synonymous with "chat" branding, as broad as that term is, it would be OpenAI with its ChatGPT models.

And if you're wondering why a not-for-profit company would care so much about branding: don't. It's now not even an open secret so much as an open fact that OpenAI is vying to become for-profit. Just a few days ago the company began discussing the possibility with the California attorney general's office.

I'm a little skeptical of some recent talk about ChatGPT having already done a lot of this rebranding, however, given that the Chat/ChatGPT website and model selector still says ChatGPT, not "OpenAI Chat" or just "Chat". Plus, "ChatGPT" is such a ubiquitous name, now, that I wonder how beneficial it would be to switch away from the branding.

Then again, if AI chat really does become a new standard for interfacing with all our tech, it might make sense to pivot towards a more general branding for the long term.

It instead looks like we might have to get used to re-brands, new-brands, brands-within-brands, whatever you want to call it, as there's talk of Copilot being subsumed under the naming "Windows Intelligence". This comes from X user teroalhonen (via Windows Central), who shows references to the new naming in the appprivacy.adml file.

Windows Intelligence branding, as well as being previously used by Microsoft, as TechRadar points out, for cybersecurity upgrades, is of course reminiscent of Apple Intelligence. It might therefore make for more recognisable branding than "Copilot".

That's not to say that Windows Intelligence would necessarily replace Copilot branding, though. More likely is that Windows Intelligence subsumes Copilot, given the latter won't constitute all of the AI functionality Windows has to offer. Recall is still on the way, after all, even if it has recently been delayed even further. This, in the same way that Apple Intelligence covers all kinds of things, from Siri to AI-aided photo editing.

It's understandable that branding isn't worked out yet, of course. All these AI integrations are still in their infancy in the grand scheme of things, given we're only just now seeing mainstream desktops and laptops featuring NPUs capable of being designated an AI PC—in Windows' case, a "Copilot PC", or, perhaps now a "Windows Intelligence" PC?

I suppose if the "Intelligence" branding comes to fruition and sticks, it might make things easier. You could talk about whether you have an Intelligent computer or not, whether that's MacOS or Windows. "Is your PC intelligent?" and so on.

Myself, though? Call me unintelligent but I'm staving off all this AI malarkey for as long as possible. Update Copilot all you want, I'll be waiting until I can't bury my head in the sand any longer. Hopefully my silicon won't be sentient by then.

"This is not a CrowdStrike issue. An issue in Windows 11 24H2 is causing some specific applications to hang. CrowdStrike identified and reported the issue with Windows 11 24H2 to Microsoft and are waiting for them on a resolution."

Microsoft says it is collaborating with Crowdstrike on a solution.

Original story: Windows 11's latest build, 24H2, is officially here and rolling out gradually. Microsoft calls it a "full OS swap", so it's no surprise that there have been a fair few kinks to work out. The latest one of these, however, is with none other than CrowdStrike's Falcon sensor, the security software responsible for the worldwide IT outage back in July.

According to Microsoft (via PC World), "after installing Windows 11, version 24H2, first and third-party apps might stop responding when antimalware security solutions enable certain features. Office applications such as Word and Excel might be affected."

Although Microsoft says the problem is with "antimalware security solutions", it clarifies that "affected devices have CrowdStrike’s Falcon sensor software" and a particular policy setting enabled.

To help fix the issue for the time being, according to Microsoft, CrowdStrike has "temporarily disabled the Enhanced Exploitation Visibility Prevention Policy setting on hosts running Windows 11, version 24H2".

In the longer term, Microsoft says the two companies are "collaborating on a solution".

Prior to this temporary fix, organisations using CrowdStrike Falcon as their cybersecurity solution were apparently facing crashes in Office applications such as Word or Excel if their systems upgraded to 24H2.

I say "organisations" rather than "home users" because CrowdStrike Falcon is primarily for organisations, so most home users' won't need to worry about their personal devices being affected. It's a cloud-based antimalware solution that monitors systems and networks for possible malicious activity—an Endpoint Detection Solution (EDS).

Earlier in the year, an update for the Falcon software, which was passed over by a buggy Content Validator, caused out-of-bounds memory reads on Windows machines and ultimately resulted in BSODs on computer systems across the world. It was one weak link in the chain that led to chaos.

Thankfully, this latest issue is nothing like that: the 2H24 update has led to crashes in just some Office programs when paired with Crowdstrike's sensor, and does already have a temporary fix. Crowdstrike is also keen to suggest the issue is not its fault this time.

What we do know is that 24H2's roll-out hasn't been completely smooth. And despite it having some neat but non-revolutionary features—File Explorer file dragging improvements and a Wi-Fi refresh button, anyone?—unless I'm someone who's going all-in on the AI PC hype I'm probably sticking with 23H2 for now.

That's just me, though. Those of you more adventurous than I can have at it—well, as soon as the update rolls out for your device and location, that is.

There's a timeline attached to each recording, along which users can add their own markers on-the-go so that important or interesting or controversial moments can be easily watched later. Developers will be able to implement "game-added" markers too, meaning important milestones in a game can be easily earmarked for later viewing. It recalls the way Xbox and PlayStation consoles automatically screenshot every achievement / trophy earned.

Recording modes include background and on-demand: the former will record to a user-specified duration or storage limit, while the latter is toggled off and on at will. Replay functionality is easily accessible in the Steam overlay for super-quick reference, and of course, you'll be able to clip and crop to your heart's desire and easily share with other Steam users (or export and send further afield as an .mp4, if you want).

Since this functionality works via the Steam overlay you can use it for every Steam game, and the functionality is already fully integrated with the Steam Deck. You can even use it for non-Steam games. Conveniently, the recording only captures the relevant game window, so your desktop (and thus your well-hidden obsession with troll dolls) won't be revealed.

Is this, as Andy suggested back in June when the beta started, a death knell for Nvidia Shadowplay? Well, it will be for me. I'm not using two programs when one works just fine. I still do miss Fraps, though. Maybe PC gaming is getting too easy?

This conversation around disinformation ended up on Meta's practices because Pamela San Martín, Co-chair of the Oversight Board for Meta, was one of the key speakers.

The Oversight, according to its own FAQ, "is a body of experts from around the world that exercises independent judgment and makes binding decisions on what content should be allowed on Facebook and Instagram".

However, just a few questions down the page, it declares that the board is funded directly by Meta, with $280 million in funding over the last five years alone. This declaration of independence, when paired with the knowledge of funding, implies a tension that the other members of the panel picked up.

San Martín, whilst acknowledging the problems of AI and Meta's own need to learn from it, praised AI as a tool for battling AI misinformation.

"Most social media content is moderated by automation and automation uses AI, either to flag certain content to be reviewed by humans, or to flag certain content to be actioned."

Off the back of this, she also suggested that the best way to combat disinformation isn't always to remove it, but sometimes to inform or label it correctly. Think of the X community notes function and you have a good idea of what that looks like. She also noted that public reports of disinformation are mostly a good tool for public figures and information, and do little to dissuade harm to private individuals.

She saw pushback when talking about regulation, specifically self-regulation of boards.

"Regulation is necessary. I'm very concerned when it's speech-related, but I'm completely for regulation when it has to do with transparency and accountability" San Martín told the group.

Brandie Nonnecke, the founding director of the CITRIS Policy Lab, responded to this claim with "I don't think these transparency reports really do anything".

The argument here is that, with the amount of AI disinformation out there, a report can show thousands of actioned examples without giving a broader understanding of what disinformation is left untouched. They can give "a false sense that they are actually doing due diligence". When those reports are created internally, it can also be hard to judge the intent of the report.

Imran Ahmed, the CEO of the Center for Countering Digital Hate(CCDC), is also critical of not only Meta but Meta's Oversight Board and its incentives.

"Self-regulation is not regulation, because the oversight board itself cannot answer the five fundamental questions you should always ask someone who has power. What power do you have, who gave you that power, in whose interests do you wield that power, to whom are you accountable, and how do we get rid of you if you're not doing a good job? If the answer to every single one of those questions is (Meta) you are not any sort of check and balance, you are merely a bit of PR spin."

This is an important point when talking about regulation, and one San Martín rebuffed by noting that she cannot be fired by Meta for her reports.

As noted by session moderator, Kyle Wiggers, the Meta oversight board saw layoffs in April this year, though she assures the panel the funding for her work is overseen by a trust which has been given money from Meta irrevocably.

It can choose to not extend the terms of employees in this fund so, while they can't be fired, they can stop getting funding, and this touches on some of the weariness around transparency reports and self-regulation.

Meta's approach to AI has seen wide distrust, as can be shown by the Goodbye Meta AI chain mail, and self-regulation may not be the best way to tackle the misuse of AI.

Nonnecke suggests that transparency reports can, ironically, obfuscate the problems the report intends to tackle and questioning Meta's incentives to regulate itself feels like a necessary step in gaining a more intelligent and safe approach to AI on its platforms.