After an extended period of wrangling, the US Congress passed a bill calling for TikTok's sale or block in March, sending it to the Senate, which voted to pass the bill on April 22. One day later, US President Joe Biden signed the bill into law. 

Known formally as the Protecting Americans From Foreign Adversary Controlled Applications Act and rolled into the much larger HR 815, a bill primarily about providing military aid to Ukraine, Israel, and Taiwan, the law essentially forbids the distribution, maintenance, or updating of an application under the control of a "foreign adversary," in this case China, and more specifically ByteDance and TikTok.

In a statement posted to Twitter, TikTok said the new law is unconstitutional and confirmed that it will challenge it in US courts. "The fact is, we have invested billions of dollars to keep US data safe and our platform free from outside influence and manipulation," TikTok said. "This ban would devastate seven million businesses and silence 170 million Americans.

"As we continue to challenge this unconstitutional ban, we will continue investing and innovating to ensure TikTok remains a space where Americans of all walks of life can safely come to share their experiences, find joy, and be inspired."

In a TikTok video, Shou expressed similar sentiments and also claimed that the real goal of some US lawmakers is not to disconnect China from TikTok, but nothing less than an outright ban on the platform, presumably a more alarming potential outcome for people who use it.

"That will take TikTok away from you and 170 million Americans who find community and connection on our platform," Shou said.

"This is actually ironic, because the freedom of expression on TikTok reflects the same American values that make the United States a beacon of freedom. TikTok gives everyday Americans a powerful way to be seen and heard, and that's why so many people have made TikTok part of their daily lives. Rest assured, we aren't going anywhere. We are confident and we will keep fighting for your rights in the courts."

@tiktok

Response to TikTok Ban Bill

♬ original sound - TikTok

It's ironic to hear the CEO of a Chinese company glorifying the virtues of American freedom of expression; it's also ironic that the US presents itself as hellbent on protecting American privacy from the predations of TikTok while allowing Twitter and Facebook to run wild. Privacy concerns are valid but also almost entirely moot in a world where the tendrils of social media have pried into virtually every aspect of our lives, and while I certainly don't consider TikTok a force of pure altruism, efforts to shut it down are clearly driven by politics above all else.

None of which makes Shou wrong: Organizations including the ACLU, the Electronic Frontier Foundation, Fight for the Future, and PEN America have spoken out against the law, saying it will "violate the First Amendment rights of Americans across the country" and "set an alarming global precedent for excessive government control over social media platforms."

Despite the president signing the bill into law, what happens next remains to be seen. ByteDance has a year to offload TikTok, and assuming it follows through on its promise to take the whole thing to court, that'll delay a resolution even further. The potential election of Donald Trump to the US presidency could also complicate matters: Despite trying to impose his own ban on TikTok in 2020, Trump has now come out strongly against the proposal and might try to undo, or at least weaken, this new law. 

China's government, meanwhile, has expressed strong opposition to a forced sale of TikTok, and more specifically the algorithms that have driven its success. For the moment, that leaves ByteDance little room to maneuver, but it needs to do something: Other Western countries including the UK, Australia, and Canada are considering TikTok bans of their own, and will almost certainly follow along whatever path the US decides to take.

Just a few years ago, anything created via generative AI was instantly identifiable, by several factors. With still images, it would be things like the number of fingers on a person's hand or even just something as simple as having the correct number of legs. AI-generated video was even worse, but at least it was very meme-worthy.

However, a research report from Microsoft shows that the obvious nature of generative AI is rapidly going to disappear. VASA-1 is a machine learning model that turns a single static image of a person's face into a short, realistic video, through the use of a speech audio track. The model examines the sound's changes in tone and pace and then creates a sequence of new images where the face is altered to match the speech.

I'm not doing it any justice with that description, because some of the examples posted by Microsoft are startlingly good. Others aren't so hot, though, and it's clear that the researchers selected the best examples to showcase what they've achieved. In particular, a short video demonstrating the use of the model in real-time highlights that it still has a long way to go before it becomes impossible to distinguish real reality from computer-generated reality.

But even so, the fact that this was all done on a desktop PC, albeit one using an RTX 4090, rather than a massive supercomputer shows that with access to such software, pretty much anyone could use generative AI to create a flawless deepfake. The researchers acknowledge this in the research report.

"It is not intended to create content that is used to mislead or deceive. However, like other related content generation techniques, it could still potentially be misused for impersonating humans. We are opposed to any behavior to create misleading or harmful contents of real persons, and are interested in applying our technique for advancing forgery detection."

This is probably why Microsoft's research remains behind closed doors right now. That said, I can't imagine it will be long before someone manages to not only replicate the work but improve it, and potentially use it for some nefarious purpose. On the other hand, if VASA-1 can be used to detect deepfakes and it could be implemented in the form of a simple desktop application, then this would be a big step forward—or rather, a step away from a world where AI dooms us all. Yay!

Now, Microsoft is going to more aggressively push Windows 10 users to use an online account too. Our sister site Windows Central reports that Microsoft is going to start pushing notifications to get users to ditch local accounts on Windows 10 in favor of a Microsoft account. The change is mentioned in the release notes for the Windows 10 19045.4353 preview build under the section titled 'features and improvements'. A public rollout is scheduled during May.

The specific text says: "A Microsoft account connects Windows to your Microsoft apps. The account also backs up all your data and helps you to manage your subscriptions. You can also add extra security steps to keep you from being locked out of your account."

If that was all a Microsoft account did, then I'd have little problem with that, but I'm sorry, I view my personal computers as exactly that—personal. I don't want to give the world's most valuable company access to my data to do whatever the hell it likes.

Thankfully, Microsoft won't lock you out of your local account should you opt against making the switch. It'll just nag you to do it. You can turn off the notifications by going to Settings > Privacy & security > General and then deselecting everything.

It's clear that Windows 10 isn't going anywhere. If Microsoft can't get users to upgrade, then it's having to find other ways to get Windows 10 holdouts to become part of the Microsoft data machine.

If you are thinking of making the switch to Windows 11, the good news is that its not difficult to bypass the requirement for an online account. Just follow this handy little guide courtesy of our very own Andy Edser.

Installing Windows 11 without creating a Microsoft Account

When setting up a Windows 11 PC using the Out-of-box experience:

1. When presented with the "Is this the right country or region" screen, press Shift + F10 to open a command prompt window.

2. Select the window and type: oobe\bypassnro (no spaces) followed by Enter.

This will cause the computer to restart.

3. Once back at the same screen, press Shift + F10 to open the command prompt once more, and this time type: ipconfig /release (again, no spaces)

4. Now press Enter and type exit to close the window.

5. Now make your way through the usual setup process until you reach the "Let's connect you to a network" screen, select "I don't have internet" and then "Continue with limited setup".

When performing a clean Windows 11 install from USB:

Create a bootable Windows 11 USB install drive using the RUFUS USB Creator Utility, which will allow you to bypass TPM, Secure Boot, and Microsoft Account requirements.

Congratulations! You should now be able to set up the rest of your Windows installation without a Microsoft Account. 

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. 

Kill It With Fire 2

Steam‌ ‌page‌ ‌
Release:‌ April 17
Developer:‌ Casey Donnellan Games LLC

As a former arachnophobe who's now a proselytizing member of Team Spider (thanks to the charms of these things)  I can understand why people might take pleasure in a game about destroying spiders. That's what 2020's Kill It With Fire was all about, and this sequel adds online coop and PvP into the mix. It's a pretty simple concept on the surface: use spider-tracking technology to hunt down spiders and then kill said spiders, usually with fire or a gun, but also by other unconventional means, if you want. It's a bonkers first-person shooter in some ways, with maybe slight influence taken from the hidden object genre? Anyway, it's in Early Access now, with a 1.0 launch expected later this year.

共鸣 ReMix

Steam‌ ‌page‌ ‌
Release:‌ April 17
Developer:‌ GameSmith

If Crypt of the Necrodancer's genius combination of pixel art, rhythm game and roguelite pleased you, then 共鸣 ReMix will probably rock your socks. It's kinda a bullet hell dungeon crawler, except the different classes are musicians specializing in different genres, meaning you'll have to adjust to the sonic qualities of say, funk, if you want to roll out with that character. Someone in the Steam reviews described it as "Rhythm Vampire Survivor" and, well, that's me sold.

The Mildew Children

Steam‌ ‌page‌ ‌
Release:‌ April 18
Developers:‌ The Growing Stones

The Mildew Children brings point 'n' click adventure and visual novel elements to a story about scary children. Sure, most children are scary, but the ones in The Mildew Children follow "savage pagan traditions"—and presumably not in pursuit of Robux. You can chat to some of these children, which will lead to some pretty heated conversations, I'm sure, since the protagonist is a witch sent to presumably liberate the village from these diminutive miscreants. The art style is painterly and slightly eerie, and your decisions, in case it wasn't obvious, will matter.

Vivat Slovakia

Steam‌ ‌page‌ ‌
Release:‌ April 18
Developer:‌ Team Vivat

Vivat Slovakia is an ambitious open world action game set in 1990s Bratislava, shortly after the dissolution of the Soviet Union and the Czechoslovak Federation. As a secret police officer you'll undertake missions that are—and here's what appeals to me—"based on real events set in the '90s of the young Slovak Republic". So it's partially a historical effort, and that's supported by the vehicles you'll drive too. It's developed by a Slovakia-based studio, and while some players are reporting performance issues at present, that's kinda par for the course with Early Access. It'll launch into 1.0 later this year with, among other things, a dog companion.

Jo 'n Jo

Steam‌ ‌page‌ ‌
Release:‌ April 19
Developer:‌ Ethan Robichaux, Dean Goodson

That's the beauty of PC gaming in the 21st century: this game about two giraffes tied together and forced to complete a terrifying obstacle course in a world of hyper-exaggerated ragdoll physics can release to almost no fanfare, because people are just so used to how many great games there are! This is free and looks really stupidly fun, in the same way QWOP, or Getting Over It, or Jump King, or last week's The Game of Sisyphus are "fun". It's also free, so no excuses.

Thanks to the release of a new version of Tiny11 builder, it's possible to create a de-bloated Windows 11 image from any Windows 11 release, allowing you to free up system resources for the apps you actually want to use.

Tiny11 Builder is a free tool created by NTDEV. It's been around for some time, but was limited to specific Windows builds. The new version (spotted by Neowin) allows you to use it with any Windows 11 version, as well as any language or architecture. This ability to take control of your Windows comes as a result of improved PowerShell capabilities. So, thanks for that Microsoft!

Tiny11 is available at NTDEV's GitHub page. To run it, you need to set PowerShell to unrestricted using the Set-ExecutionPolicy unrestricted command. Once that's done, just follow the instructions provided.

Tiny11 removes the following apps: Clipchamp, News, Weather, Xbox (although Xbox Identity provider is still here, so it should be possible to be reinstalled with no issues), GetHelp, GetStarted, Office Hub, Solitaire, PeopleApp, PowerAutomate, ToDo, Alarms, Mail and Calendar, Feedback Hub, Maps, Sound Recorder, Your Phone, Media Player, QuickAssist, Internet Explorer, Tablet PC Math, Edge and OneDrive.

For good or evil, I use Google for browsing, mail and work tasks, so as far as I'm concerned, I don't need any of the stuff that gets removed. The one exception is Edge, which I would normally use to download Chrome or use it as a backup browser in case a page glitches in Chrome, though that's easy enough to grab from the Microsoft Store.

In addition to removing those apps, the tool also bypasses the requirement to create or use a Microsoft account. You could skip that using a tool such as Rufus already, but it's nice to have Tiny11 remove it as well.

I took the new Tiny11 builder for a quick spin using an older Windows 11 iso I had on my work PC. After running the script, the resulting iso file was a full gigabyte smaller than the original.

I have an older Dell XPS laptop that doesn't meet the system requirements for Windows 11. I plan to use the Tiny11 iso to give it a new lease of life, free of the TPM requirement. And, it's sure to use less RAM too.

One of Grok's primary purposes is to summarize breaking news on Twitter, however it has been prone to the odd bout of confusion, like getting itself into hot water by completely making up significant world events. Most recently, Grok appears to have mis-interpreted a joking Twitter post accusing a basketball player of "throwing bricks", which is commonly used in reference to a player making a shot that doesn't hit the rim (via Ars Technica). 

While Klay Thompson appears to have had a bad night on the court in what was said to be potentially his final game with the Golden State Warriors on April 16, the befuddled AI appears to have taken the tongue-in-cheek Twitter post literally in regards to a current NBA star, publishing a post entitled "Klay Thompson Accused in Bizarre Brick-Vandalism Spree".

While a disclaimer appears underneath Grok's reports informing readers that "Grok is an early feature and can make mistakes" and encouraging them to verify the AI's outputs, many had already reposted the comment, which seemed to confirm to the AI that they too had been a victim.

Access to Grok has recently been granted to all premium Twitter users, making this quite the public mis-step at a time when many might be debating whether to put any trust in the feature's news-summarising capabilities.

Klay Thompson has yet to issue a statement on the false report, although I'd imagine that Elon Musk and co. might be hoping that he sees the funny side. Despite the disclaimers and warnings that come with the AI feature, both Microsoft and OpenAI have recently had to contend with lawsuits in regards to their own AI chatbots creating false information that may be potentially libellous.

While its easy to poke fun at an AI going off the rails, underneath here does lie a serious point. With the continuing popularity of AI chatbots, image generators, and more, it seems to be increasingly difficult for many to tell real content from an AI-generated fake.

While this one was particularly easy to spot, it does make you wonder as to how much information we take for granted may, at its root, have come from erroneous AI generation.

After all, if the story summation had read "Klay Thompson has the best basketball game of his career", many, myself included, may have taken it at face value.

For the record though, just in case Grok wants to summarise this story once it hits Twitter, I'd like it to read "PC Gamer hardware writer wears flowerpot as a hat". That should generate a headline I can really be proud of.

That's according to the 2024 Bad Bot Report by cybersecurity specialist Imperva, via SecurityBrief. The headline figure has bots representing 49.6% of all internet traffic. Arguably even worse is the 32% of all traffic associated with so called "bad bots."

"Bots are one of the most pervasive and growing threats facing every industry. From simple web scraping to malicious account takeover, spam, and denial of service, bots negatively impact an organisation's bottom line," Nanhi Singh, General Manager, Application Security at Imperva, said, presumably with an apathetic, defeated sigh.

Imperva found that bad bot traffic varied fairly dramatically by geography. For instance, Ireland topped the charts at 71%, with Germany at 67.5% and and Mexico notching up 42.8%. It's not clear if that means 71% of all Irish internet traffic is bad bots or rather if 71% of Irish bot traffic is bad. To be sure (apologies), that's a lot of bad bots.

What's more, we gamers were not left out of the fun. In fact, according to Imperva the prevalence of bot traffic within gaming was deemed the greatest among all "sectors" at 57.2% of traffic. Go us! That said, Imperva didn't break down the good-to-bad bot ratio for gaming.

Retail bots achieved a mere 24.4% of traffic, with travel bots representing 20.7% of that sector's traffic and financial services relegated to the must-try-harder-to-assimilate-humans category at just 15.7% of traffic.

Imperva further reckons that "advanced" bad bots, that are said to be capable of mimicking human behaviour and evading defences, were most common on Law and Government, Entertainment, and Financial Services websites.

All of which means that the overarching narrative here is a plague of bots on all our houses. "As more AI-enabled tools are introduced, bots will become omnipresent. Organizations must invest in bot management and API security tools to manage the threat from malicious, automated traffic," Imperva's Nanhi Singh added.

It's all pretty depressing when you consider the energy footprint of the internet as a whole and therefore all the megawatt hours guzzle up by bad bots chasing us and each other around the internet and training each other on terrible AI-generated content.

Exactly what can be done about any of this is an open question. It's tempting, rather ironically, to appeal to AI as a potential solution to all the bad bots and nonsense content. But that does seem rather circular and zero sum, a battle of AI wits that feels like it can only end one way. Yup, that post-apocalyptic wasteland and a pile of skulls.

Or maybe that's a little pessimistic and all we have to do is look after our passwords a little better, not click on dodgy links, and take it easy on the social. Fight it out about the dead internet theory in the comments below.

For a while, I doubted myself: Maybe I had told Windows to "update and restart" instead of "update and shut down." And maybe I should go check the burners on the stove, because clearly I can't be trusted with on/off states.

But no, it wasn't me. When I mentioned this experience in a meeting recently, another PC Gamer editor said, Wait a minute, you too? 

Windows has been sporadically ignoring the "restart and shut down" option since at least 2021, when someone asked what was up with their PC restarting instead of shutting down on the official Windows forum. Then, about eight months ago, a dozen or so people reported the problem in a Reddit thread. That's around the time it started happening for me.

"Because of this weird bug, I can't trust selecting 'Update and Shut Down' at the end of my work day," wrote the Reddit poster. "It works around 50% of the time, but often the system would restart 2-3 times to finish applying update and end up at the login screen. If I leave my laptop on battery power, it would be dead when I find it the next morning."

Some say the bug occurs whenever Windows needs to restart multiple times to apply an update, but I'm not sure about that—for me, it seems like it just restarts once, regardless of whether I tell it to update and shut down or update and restart. Back in that 2021 thread, someone said it was a BIOS issue, so you might try updating your motherboard's firmware, but I can't say for sure whether or not that's the problem.

Funnily, if you go back far enough in the Google search results you can find someone complaining about the "update and shut down" feature doing what it says it does. Rather than restarting to finish the update, it was leaving the PC shut down and finishing the update on the next boot.

Well, fair enough—that's not ideal, either. The best version of the system would reboot as many times as needed and then 'remember' to shut down as requested. That seems to be how it's meant to work in Windows 11. But it doesn't.

I can't find any official Microsoft support comment on the matter, but on the chance that it's my fault for being lazy about motherboard maintenance, I'll refrain from ranting about long-standing Windows bugs here. (Though I really can't guess why a slightly old BIOS would have anything to do with this, and it probably doesn't.)

Mainly, I just wanted to let anyone who's experienced this know that they haven't lost their grip on reality. You did lock your car. You did turn off the oven. And you did click "update and shut down."

I'll confess to being rather partial to Logitech's mice, namely the MX Master series and my current choice of rodent, the MX Vertical. However, I've never bothered to dig into what the Logi Options+ app has to offer, mostly because it's never really had anything worth digging into. That could well change with the introduction of the AI Prompt Builder feature in the latest 1.7 version of the software, which was released today (via The Verge).

Essentially, it's a simplified interface for using ChatGPT 3.5, or if you have a Pro account with OpenAI, ChatGPT 4. By default, the system offers the choice of rephrasing or summarising a body of text, along with the ability to generate a reply to something or create a new email. That's no different to the numerous other AI systems that are starting to embed themselves into every aspect of a modern PC, but at least it's very unintrusive.

During the AI Prompt Builder setup process, you're forced to assign the tool to one of the buttons on your Logitech mouse. Not all models will let you do this, but I had no issues with my MX Vertical and I chose the 'cursor speed' switch on the top of the mouse, as it never gets used. I have to say that it all worked rather nicely, though if you've ever used ChatGPT before, you'll know exactly what to expect concerning the responses it generates (i.e. it's pretty variable in terms of quality and accuracy).

If you like this idea and don't have a mouse with a suitable button, then fret not as Logitech is launching at least one mouse that does—the Signature AI Edition M750 wireless. At $50/£55, it seems a tad pricey for an ordinary mouse with a pre-configured button to launch AI Prompt Builder when its Signature M650 model is $10/£10 cheaper.

The software feature is pretty nice, with a clean and relatively minimal interface, and for those users who use ChatGPT a lot, it's a great way to quickly jump into the system whenever you need it. On the other hand, shoehorning one more button onto a standard mouse and charging you an extra 10 notes for the privilege is a bit cheeky, in my opinion.

At the moment, we're in the calm before the storm, and it won't be long before there's an absolute flood of AI-buttoned mice, keyboards, gamepads, headsets, and other peripherals all offering the same feature. Because it's the year of the AI PC, in case you weren't aware.

The bots are really the most notable thing because, while Twitter's always had problems, the uptick in bot activity over the last couple of years has notably impacted the site's user experience. Ironically enough, one of Musk's big reasons for buying the platform was to annihilate the bots, but under his stewardship the problem's gotten significantly worse.

The solution? Musk has returned to an idea first mooted last year, and already trialled in two countries as the "not a bot" scheme: charging new users a nominal fee (the suggestion was $1 a year, or local equivalent) before they're allowed to post anything. Accounts will still be free to create, but can only read posts.

A news account that reports purely on X noted a recent change to the site's policies which suggested that this feature is going to be rolled-out more widely. No less than Musk himself shortly waded in to confirm that, yep, soon enough people will have to pay for the privilege of posting.

"Unfortunately, a small fee for new user write access is the only way to curb the relentless onslaught of bots," said Musk. "Current AI (and troll farms) can pass 'are you a bot' with ease.

"The onslaught of fake accounts also uses up the available namespace, so many good handles are taken as a result."

I feel a bit like I'm writing about Helldivers 2, but suffice to say the war against the Automatons looks to be entering a new stage. The idea itself is not completely wild and, if implemented correctly, would undoubtedly curtail some bot activity: but to what extent may well depend on what Musk decides to charge. $1 an account may not be too much of a disincentive for crypto scammers out to promote their latest pump 'n dump scheme.

For all the exaggerated chat about Musk's regime "killing" Twitter, however, this may be the one thing that has the potential to actually do it. It's going to piss off casual users (though there's no suggestion as yet it will be applied to existing accounts) and turn a service that used to be free into a subscription, even if the cost is minimal. That's even before you get into the perception, fair or otherwise, that Twitter itself is a shadow of what it used to be (personally speaking, I find its usefulness as a news source has been almost entirely subsumed by spam and misinformation). 

More than anything, it's putting the cost of fixing this problem on the masses, and those who are less interested in following the ins and outs of big tech may well find the whole bot rationale a weaksauce excuse for what could be interpreted as a shakedown. 

X's page says the Not a Bot plan has been developed to "bolster our already successful efforts to reduce spam, manipulation of our platform and bot activity, while balancing platform accessibility with the small fee amount. It is not a profit driver. And so far, subscription options have proven to be the main solution that works at scale."

Reporting on this idea when it was first mooted last year, PCG's Andy Chalk suggested that Musk charging users "as part of a purported effort to clean up his own self-made mess is an ass-backwards approach," which is roughly where I land. It's probably worth pointing out that this has not yet been implemented on a wider scale, and Musk does tend to shoot from the hip when discussing ideas and policies in public.

But Musk has talked a big game about his plans for Twitter and, a couple of years down the road, it's not really clear that he's made anything about the platform better. It remains to be seen whether this plan can make any difference whatsoever, or if it's the final nail in the coffin for the bird site.