Have you ever watched a YouTube video and immediately screwed up your face in an attempt to discern whether it's real or AI-generated? Welcome to a daily part of my life. The good news is that YouTube has listened to user feedback, and as a result is moving AI disclosure labels for photorealistic and "meaningfully AI altered or generated" content to a "more prominent" position.

The bad news is, the labels look pretty tiny in the promo video. The AI warnings will now sit directly below the video player for long-form AI-generated videos, above the description. For short form, it's an actual overlay on the video itself. Which is indeed, a prominent position.

But yes, both versions still look like they'd be easy to miss. I suppose a big honking "THIS VIDEO IS BULL****" with red flashing warning signals on either side would be a bit much, but I'm not sure it does much to assuage fears that AI-generated videos are becoming harder to spot.

What might help a little more in this regard, though, is YouTube's other announcement—that "new internal signals" will be used to help identify AI-generated content.

YouTube creators are required to manually disclose when they use realistic AI, but I'm guessing that a fair few of them... don't. With the introduction of this tech, though, the platform says it will be able to automatically apply a label to anything with "significant photorealistic AI use."

"As this technology continues to improve, creators remain in control," says YouTube. "If a creator thinks their content was incorrectly identified as AI-generated, they can update the disclosure status in YouTube Studio.

"However, disclosures will remain permanent in a handful of cases, including: Content created using YouTube's own AI tools, like Veo or Dream Screen [and] content containing C2PA metadata indicating they were fully generative AI."

"Our goal is simple," the blog post continues. "Make it as easy as possible for creators and viewers to have the right information."

A noble goal, to be sure. However, in a world where YouTube dominates the online video market (and amid real concerns about the level of AI slop on the platform), this new labelling system feels like small potatoes. Very small potatoes indeed, come to think of it.

For the banishment of all doubt, no, Google isn't scraping the contents of your email inbox in order to train its AI models. However, that doesn't mean Gmail is free of Gemini, and sooner or later you'll need to decide your own personal boundaries with AI.

Google announced at the start of the year that Gmail would be entering the Gemini era. This ushered in AI overviews for your inbox, as well as the Help Me Write feature that allows floundering users to draft emails with AI assistance. Gemini is used to sift through all the data in your inbox in order to summarise information or give relevant insights.

These features have been available for some time now, but the presence of AI in your inbox may have caught your attention again recently due to viral posts such as this one from Shark Tank's Lori Greiner (via Forbes). What you may have heard less about is 'Personal Intelligence'.

Personal Intelligence began to roll out around the same time as Gmail's Smart features, and allows users to link Gemini to various Google apps, like Gmail, Google Photos, and YouTube. You could ask Gemini 'Hey, why can't I stop thinking about low-poly rats?' and, in theory, Gemini could look through your YouTube watch history and messages, then say something like, 'It's because you rewatched Rat Movie: Mystery of the Mayan Treasure last week after your best friend sent it to you with the following caption: Rat squad 4 lyfe.'

Much like Gmail's smart features, when personal intelligence is enabled Gemini can search through all of the personal data held by Google's apps in order to answer your conversational queries. It can also scan all of your Google photos in order to "create more relevant, personal images using Nano Banana" (via Forbes).

It's worth reiterating that 'personal Intelligence' is an opt-in featureset. You can also opt-out of Gmail's smart features fairly easily if you'd rather Gemini didn't go rifling through your inbox. First, open Gmail and click the big cog icon in the top-right corner. On the Quick settings menu, click 'See all settings.' Now, scroll down to Smart features and untick the associated box. This does mean you lose access to a number of AI-assisted features, including automatic email categories in your inbox.

In the case of both 'Personal Intelligence, and Gmail's AI features, Google assured last month that its AI isn't snaffling up any of your data. This is in line with Google's assertion last year that it's not using Gemini's Gmail integration to train its AI.

That said, no inbox is really an island. Google says in support documentation for Gemini apps, "When you interact with Gemini, summaries, excerpts, generated media, and inferences from your relevant media, emails, and files may be used to help us answer your prompts. To make our responses relevant, helpful, and high quality, we train our generative AI models off of these summaries, excerpts, generated media, and inferences."

In other words, Gemini won't steal your photos or emails for training, but it will retain the conversations you've had about them and elements of that data may then be used to train the model. Even with this degree of separation, I still wouldn't want to allow any AI model to rifle through my personal data—especially not to create personalised, AI-generated memes based on my phone's photos. My digital memories are worth more than that—or, at the very least, they're worth an authentically terrible Photoshop attempt.

In a sublime example of the internet's "I made this" effect, an Italian TV channel temporarily took down Nvidia's March 16 video "Announcing Nvidia DLSS 5" via copyright strike, claiming the content as its own.

As reported by NikTek on X.com, the DLSS 5 reveal trailer was unavailable in Italy for more than 24 hours, with the message "This video contains content from La7, who has blocked it in your country on copyright grounds" taking its place. While we're doing a little reverse-engineering here to understand what happened, La7 seemingly used footage from DLSS 5 in a recent broadcast, then started issuing copyright claims based on that footage.

Brazen? Yes. Funny? Even more so, considering the overt distaste both players and the games industry have shown for DLSS5.

The impact extended further than Nvidia's own YouTube channel. Multiple other posters on X chimed in to share their own videos being demonetized and region blocked due to a copyright claim. Former IGN editor Destin Legarie posted about the incident on April 4, prompting a rather unhelpful response from YouTube's official X account explaining the basics of its Content ID system.

"The channel making these claims does not own the rights to an Nvidia trailer they used without permission during their talk show," Legarie shot back. "They just created work for YouTube by abusing the system."

Legarie posted on Monday that La7 has released the claim. Using a VPN, PC Gamer confirmed that Nvidia's reveal trailer is also now once again viewable in Italy.

YouTube's Content ID system is often criticized for being abusable, with frivolous claims on videos that should fall under fair use often meaning smaller channels see their videos demonetized unless they can contest a claim, a process that can take weeks. It's equally common for these automated systems to go too far, ending up in claims that are quickly revoked as soon as a real human being takes notice.

It's funnier to think of someone at TV channel La7 deciding it owns all coverage of Nvidia DLSS 5 after producing one talk show segment about it, but it seems more likely that someone clicked a "protect our copyright" button without fully understanding what it would entail.

Or maybe La7 was gung ho about taking credit for the new technology until it realized it was picking a fight with Dave Oshry, at which point the only safe move was immediate surrender.

Norman Caruso, better known online as The Gaming Historian, has spent the last decade and a half making Ken Burns-style documentary videos for his YouTube audience; he currently has over 1 million subscribers. He's covered everything from a TV with a Super Famicom built into it to the Sega Mega Modem. But now he's calling it quits.

A video went up on Caruso's channel today titled "Thanks for Watching," and he promises there's "nothing dramatic" behind his decision—he just got burnt out after making his most ambitious video yet covering The Oregon Trail, and in the two years since it released, he hasn't wanted to make more.

"I assumed that after a few months, I'd get the itch again and make a new video," Caruso said in the video. "To my surprise, that itch to make a new video never really came back. I definitely tried to make something else. I even announced it on social media. But my heart just wasn't in it, and I knew that if I ever finished that video the quality would suffer. That's when it finally clicked: I was ready to move on from The Gaming Historian."

Moving forward, Caruso will focus on the history podcast he hosts alongside his wife Kristin. But even though he's putting his YouTube channel behind him, he unveiled one last treat for fans: a shedload of court documents he scanned for a video he no longer plans to make.

The documents cover a legal case from 1982 where Universal Studios alleged the game Donkey Kong was a trademark infringement of King Kong. Caruso discussed the case briefly in his video on how various Mario characters got their names, and in that he showed off a court document which lists several other names that were considered for Donkey Kong during development of the arcade game—among them are Bill Kong, Kong Holiday, and Kong Chase. I tell you that because I think we'll agree that Kong Holiday is a much better name.

For more trivia nuggets like that, you can view the documents Caruso uploaded to the Internet Archive. He said he also provided the files to the Video Game History Foundation in his video. "Making the Gaming Historian was truly a life-changing experience," he said. "I will always cherish that chapter of my life."

Have you ever wondered what would happen if you ate thermal paste? I'm not sure, but I can advise you that you really, really shouldn't. YouTuber mryeester looks to have been having similar thoughts, as they've endeavoured to create the world's first truly edible thermal paste—and the secret ingredient is gold.

Gold is known for being an excellent conductor of heat, and being in position of a surprising amount of 24-carat gold leaf, mryeester picked it out as the edible thermal transfer element. Yes, you can eat it, as you'll find if you go to a very fancy restaurant. You'll also likely find a large bill at the end of the evening, as the stuff isn't cheap.

However, simply placing gold leaf atop a CPU heatspreader isn't really a solution, as it ignores the reason we use thermal paste on our coolers in the first place. A traditional paste fills in the microscopic gaps of air between the chip surface and the cooling plate, improving thermal conduction considerably, which is why it's the go-to for system builders everywhere.

So, a liquid bonding agent was required, and mryeester picked out everyone's favorite bee-based product, honey. The other candidates were mayonnaise, ketchup, mustard, and toothpaste, if you were wondering. Yes, this is going to get messy.

Our intrepid chef then proceeded to mix the gold and honey together on top of the chip, which was a bold move given that the former is highly conductive. Still, the end result was a shiny golden blend with plenty of the 24-carat stuff glistening inside, for maximum thermal conduction.

Squidging the mixture on top of a test chip with a small air cooler, mryeester recorded a surprisingly low 53 °C peak thermal figure from the un-named CPU, even at sustained 100% load.

Next up, a proper gaming PC with an AMD Ryzen 7 9850X3D inside. That's one of the fastest gaming CPUs money can buy, and a $500 thoroughbred race horse of a chip. Gulp. Using a conventional air cooler (with conventional thermal paste), mryeester recorded a peak temp of 70 °C while playing a match of Counter Strike 2.

That's well below the thermal limit, but what could the honey and gold mixture offer instead?

Well, less of a mess, as mryeester was quite rightly concerned about getting too much honey on their very expensive chip and motherboard CPU socket. However, the volume used was still much more than what I'd ever recommend applying to a chip with the equivalent paste. Pea-sized amounts, folks. Pea-sized amounts.

Time for a taste test. Mmm, delicious, apparently. Gold doesn't really taste of anything, whereas honey tastes of golden fields and your imaginary childhood. Or mine, at the very least.

Large air cooler dutifully installed, a last-minute panic was averted due to the PSU switch being left off during installation. Hands up how many of us have done that? Yep, me too. Still, the PC was up and running in no time, and reported peak gaming temperatures of.... 77-78 °C.

That's a fair bit toastier than the regular thermal paste, but still well within reasonable limits for a 9850X3D. At no point did the chip thermal throttle and limit performance, although as mryeester admits, that was over a 10-minute CS 2 session.

Longer playtime (or a more CPU-demanding game) would likely lead to worse and worse results, not least because non-pure honey is a (somewhat) perishable product, and will likely change in consistency over heat cycles and time.

So, what have we learned? Well, that proper thermal paste makes a significant difference to CPU temperatures, I guess. However, it's a genuine surprise to see that a relatively unsuitable product (even with gold's thermally conductive properties) doesn't appear to impair the CPU temps by all that much, and you could lick it off to clean up the mess afterwards.

Only joking. Don't lick your PC components, folks. Actually, don't do any of this. It's for YouTube entertainment value and dubious science experiments only. Keep the honey for the kitchen, and do yourself a favour and spend $10 on some proper thermal paste for your next build. It's likely a whole lot cheaper than a pack of gold leaf, anyway.

Have you ever thought, 'There aren't enough fans in my fan?' That's what I can only assume is the thought process of YouTuber Major Hardware, who has replaced one Noctua fan with 15 mini ones.

The inspiration for this project is a tiny 'flying UFO' toy that their kids were playing with. Major Hardware says, " While watching it go around, I noticed that these little tiny fans do produce quite a bit of wind. They're very loud, but... they produce quite a bit of air flow."

Major Hardware found a similar 5V motor to the one in the UFO, bought a big pack of them to plug into a sequence of Noctua colored 30 mm fans, and found that 15 would offer a similar surface area to the 110 mm one they own.

From here, they used the base of the Noctua NF-A12x25 in a 3D model, and fit a plastic dome over the top to slot the fans in. They broke a motor mount in the first model, causing cracks due to a lack of an antivibration pad in the base. This is just a handful of problems that popped up in the first attempt. The joys of DIY.

The second model fixes those issues, and also routes the motor wires around the base, so as not to interrupt airflow. Major Hardware compares it to the Manhattan Project, but the top comment astutely re-names it "The Fanhattan Project".

Major Hardware does try to make The Fanhattan Project close to Noctua's in look, saying "this fan is basically a meme of the A12x25". Once turned on, though, it really sings. Or, perhaps more accurately, screams. "It's moving quite a bit more air than I expected. It sounds like a bunch of angry bees."

If you're wondering how loud exactly, Major Hardware's fan registers 73 decibels, with nearly no peaks or troughs. That's reasonably close to the sound of a dishwasher or vacuum cleaner. However, sound isn't the only thing you'd test when setting up a fan.

In a testing suite, with an old Intel Core i7 7700K running for twenty minutes, the standard Noctua fan saw the CPU reach a temperature of 69.5 °C, where the Fanhattan Project got 69 ° C in the same tests. That's technically better, but only fractionally so, and, as pointed out by Major Hardware, "pretty much within the margin of error".

Still, if you can DIY your own biblically accurate Noctua fan and not get worse results, you're winning in my book—just excuse all the noise.

YouTuber, extreme overclocker, and Thermal Grizzly CEO Der8auer has been keeping themselves busy of late, testing the pressure resistance of water block cooling windows. Traditionally, these are made of a durable acrylic that can easily withstand far more pressure than a conventional liquid cooling system can provide, but Thermal Grizzly has been experimenting with glass windowed versions.

And by experimenting, I mean hooking them up to a hand pump in a water bath and ramping up the pressure until they go bang. Important work is being done here, folks. Important work.

Both acrylic and glass windowed blocks were placed inside a grey plastic crate, submerged in water, and then pumped to within an inch of their lives to see what sort of internal pressure they could withstand before cracking into pretty (but useless) patterns.

Der8auer explains that the test could also be performed in the open air, but risks an explosion of glass/acrylic that would be... well, bad news for everyone involved.

First up was a traditional GPU water block with a PMMA (acrylic) window. Attached to a pump mechanism and submerged safely below the surface, the block was filled with water with the opposing plug open, so as to force out any air bubbles. Once the water block was plugged, creating a closed system, an extra piece of acrylic was placed on top of the tank to prevent any unforeseen accidents.

Ramping up the pressure, the acrylic window cracked between eight to nine bars of pressure, although did not burst into fragments. This is way beyond the pressure a normal water cooling block would reach under usual usage, and rather impressively, the acrylic was able to be removed in a single piece.

As the camera wasn't on the block at the exact time it failed, Der8auer repeated the experiment, and this time cracked a corner of the window at nine bars of pressure.

Switching over to a glass-windowed version, the test was performed with the same methodology once more. The glass shattered in a much more densely-cracked pattern at around five bars of pressure. This was an expected result, as acrylic is a much more flexible and pressure-resistant material than glass, which is more brittle.

Finally, a CPU water block with a much smaller glass window was subjected to the torture test. This managed an impressive 10 bars of pressure before failure.

Your average water cooling block is subjected to far less than 0.5 bars of pressure under load, so a five bar fracturing result is still more than good enough by most standards. And, as der8auer points out, at these sorts of pressures, it's far more likely that a hose, pump, or clamp would fail before the windows did, glass or otherwise.

"Honestly, we're just also having a lot of fun playing around with these things. I absolutely enjoy that," says Der8auer. No kidding. Can I get an invite to the next "blowing hardware up" party, please? I'll bring snacks, I promise.

Original story:

YouTube is struggling at the moment. Although I can still play videos and load some pages, attempting to browse the video sharing site returns "something went wrong" errors for me.

DownDetector has recorded a huge spike in reports—340,000 at the peak—so if you're also experiencing problems, we're definitely not alone. YouTube has confirmed that the issue is being looked at.

"We're aware some of you are having issues accessing YouTube right now," reads a YouTube support page. "Our teams are aware, and we'll provide updates as soon as we have them."

YouTube elaborates: "An issue with our recommendations system prevented videos from appearing across surfaces on YouTube (including the homepage, the YouTube app, YouTube Music and YouTube Kids). The homepage is back, but we're still working on a full fix – more coming soon!"

We'll update this article when we learn more, or if the issue is resolved. In the meantime, you can enjoy the comments on that support page, such as "I was going to draw while playing music. I pay for premium btw so I need it like now." Fair!

Google and Alphabet CEO Sundar Pichai has handily summarised the latter's Q4 results in a blog post this week, and right near the top is an interesting figure: YouTube's annual revenues surpassed $60 billion across ads and subscriptions.

Looking deeper into the fiscal results themselves, YouTube ads brought in $11.38 billion in the last quarter, compared to $10.47 billion in last year's report. Anecdotally, I can well believe this figure, as YouTube's ad frequency ramped up at such a rate towards the end of 2025 that I finally caved and bought a YouTube Premium subscription myself.

Speaking of which, while Alphabet doesn't report YouTube Premium's revenue individually, it has declared that the company now has "over 325 million paid subscriptions across consumer services, led by strong adoption for Google One and YouTube Premium."

On the surface, YouTube's revenue figure seems impressive—but its viewing figures are even more so. According to CEO Neal Mohan in his 2026 YouTube community letter, YouTube Shorts alone average out to 200 billion daily views. The YouTube head honcho also reports that YouTube has been "#1 in streaming watchtime in the U.S. for nearly three years", according to Nielsen's The Gauge report.

It's difficult to get reliable data on YouTube's usage figures overall, but it's said to have between 2.5 and 2.7 billion monthly active users worldwide as of 2025, with well over 1 billion hours of video content consumed daily, and over 20 million videos uploaded on average per day. If those figures are even close to accurate, that's an astonishing amount of data to handle—even for a cloud behemoth like Google with its own vast server infrastructure to lean on.

Likewise, it's very difficult to get a bead on exactly how much YouTube costs to run. But once creator monetisation payments, employee costs, and server maintenance are brought into the equation, it must be a pretty penny. Alphabet's Q4 financial report says that the revenue of Google Services in total was $95.86 billion, with its operating income reported as $40.132 billion, so that's $55.73 billion spent on servicing costs in the last quarter alone.

Of course, that's including Google's vast services infrastructure overall, not just YouTube. Still, it does make me wonder whether YouTube in and of itself is as profitable as its parent company hopes it would be, despite the impressive revenue figure. It's certainly more than Netflix, though, which reported $45.18 billion of total annual revenue earlier this year—along with operating expenses of $31.85 billion.

The ad strategy I (and many others) have observed in recent months certainly seems to suggest that YouTube is doubling down on its income potential, whether that's advertising or pushing people towards its paid premium services, or a combination of the two. It's remarkably good at noticing adblockers these days, too, which feels like an escalation of its previous, adblock-hating ways.

It's all about money at the end of the day. Still, with so many eyeballs on its videos, and so many excellent (and many dubious) creators using its platform, I will admit that YouTube Premium has finally got me to open my moth-ridden wallet and cough up. What can I say, I like a bit of ad-free YouTube before bedtime. Don't we all?

The monsters in Akalabeth: World of Doom—the precursor to the Ultima series, sometimes called "Ultima 0" by fans and series creator Richard Garriott himself—look ridiculous. The thief is just a floating cloak, the mimic is a featureless cube, and the final monster (straight-up called a Balrog) looks a bit like Firebrand if he was crushed under a cartoon steamroller. To be fair, it was 1979 and Garriott was a teenager; the Apple 2 could only draw a pittance of lines on screen at a time, so he designed wireframe silhouettes for each enemy using coordinates on graph paper.

Akalabeth's world was a meager, primitive trick of the light. But in the days of text-based multi-user dungeons and Zork it was a crumb of revelatory proof that the emergent worlds players imagined in freeform sessions of Dungeons & Dragons—which was only five years old at this point—could be cast in a virtual mould, simulated with math, and explored through the phosphor glow of a CRT monitor.

YouTube documentarian Majuular, or Luke, was not at ground zero for this CRPG revolution. He grew up with the RPGs that live in Ultima's long shadow and had been making videos sifting through their complexities since age 14, informed by the zany YouTube stylings of the Angry Video Game Nerd and Classic Game Room. But while he was researching the origins of his most beloved game genre and reading the 450-page tome Dungeons and Desktops: The History of Computer Role-Playing Games, he was singularly struck by Akalabeth's futurism.

"Something broke through in my head and I realized how incredible it was that somebody thought this up, made it, and put it out, and not solely Akalabeth," he recalled, measuring out each word with the enthusiasm and clarity of a habitual storyteller. "It was like the root of a tree that grew out and became this incredibly nuanced world of RPGs that I already had grown up with. I had always taken that seedling for granted."

Galvanized by Barton's book, Luke was inspired not only to play through the series which he had only dabbled in previously for himself, but to use his newest, relatively young YouTube channel to relay the wild story of its creation in his own way. Charting the sparse corridors of the World of Doom to the endless eccentricities of creator Garriott (represented in each game as the preposterously named Lord British, ruler of Britannia) has been a two year undertaking, but more than 500,000 people are showing up for each new video.

Britannia makes the waves

Eleven videos deep as of late 2025, the Majuular Ultima Retrospective embodies the fastidious, affectionate critique that is gaming YouTube at its best. Many are multi-hour features that trace Ultima's burgeoning influence on both western and Japanese roleplaying games, discuss the human stories and context around its creation, and actually involve the dirty work of playing the games while analyzing Ultima's lore on its own terms.

While you might associate gaming-centric video essays with long-winded synopses or provocative hot takes, Luke's Ultima series is the fruit of exhaustive research compiled from decades of interview material, elusive photographs, and details cataloged by communities like the Ultima Dragons virtual fan club (without whom Luke says he'd be "totally lost").

That research might seem tiring given the pace at which Luke has pumped out these videos while also making detours to the likes of William Shatner's TekWar—he's covered 1979's Akalabeth through 1993's Ultima 7 Part Two in just two years—but immersing himself in each game's history is precisely what keeps him going.

A single question remains his creative north star: "What would it have been like to dream this up at the time?"

"There's something really intriguing about knowing there's just a group of young guys with very little experience in the real world who are actually able to create something that is a real contender, something that can threaten these huge videogame companies that have infinitely more employees and resources. It turns out that just a good idea and infinite drive and lots of Diet Pepsi is pretty good fuel for redefining entire genres or even creating them, in the case of CRPGs or id with the FPS," he said.

"You almost wish you were there. You wish you could hang out with these guys and spend the afternoon eating leftover pizza in [early id Software's] lake house … there's something romantic about that kind of scrappy, amateur, just-doing-it-for-the-love-of-it and hoping something comes out of it, then not only does something come out of it, but something that is still very obviously prevalent to this day. You cannot ignore the influence of these people and the things that they've created."

Exodus

Luke has over 200,000 subscribers on YouTube and a healthy following on Patreon these days, but he's been making videos for as long as he's had the means. His earliest works were "angry reviews," ripoffs of AVGN. He didn’t pull in many viewers then, but his was a wildfire enthusiasm for games: crude, self-sustaining, and uncontainable.

"It turns out that taking somebody's shtick and doing the same thing but worse usually doesn't bear fruit on YouTube," he says, but those attempts taught him how to write a script and cut together gameplay clips.

Before Majuular, Luke worked for the better part of a decade as a pizza delivery driver, a retail worker, and eventually in manual labor to make rent as he moved with his girlfriend "from basement to basement" as a college dropout. He noticed YouTube video essays become a huge part of the online conversation as channels like Joseph Anderson, MandaloreGaming, and Matthewmatosis rose to prominence, but it all seemed like watching the back of a train he'd missed slowly become a pinhole on the horizon.

"What a great idea, to talk about videogames in this way," he remembered thinking. "Well, it's too late. Somebody else has already cracked the code; back to the lumber yard with me."

It was in a rare clump of obligation-free months between a second college exit and a planned return to hauling construction material that he decided "to go all-in" on a few highfalutin dreams. One of those dreams was still YouTube, and so he pumped out lengthy retrospectives on the likes of Chrono Cross, The Legend of Dragoon, and eventually, Ultima, on a new channel he named after a Devin Townsend song. Little by little Patreon members trickled in.

Now with 2,700 people paying to see more, Luke and his girlfriend have saved up enough money to afford "a home, and sunlight for our cats." It's the first time they've had a kitchen and windows in years.

"It's totally changed my life," he said. "It's given us a degree of security and it's just cool to have a group of people who are interested in what you have to say and it's something I don't think I'll ever get used to. Seeing people say 'I'm so excited for this video' … it's almost overwhelming."

His work ethic has hardly changed since his days at the lumber yard. "Very much a creature of habit," Luke says—he puts in his hours like clockwork each day, whether they're spent researching, playing and note-taking, or trawling obscure archives for an era-appropriate picture of Warren Spector. It's natural to spend a dozen hours or more a day working on the next video, split between script writing (which he “rarely” looks forward to, dreading the piercing gaze of a blank Word document) and the "more workmanlike," if tedious, editing process.

The routine is fueled by the sense that he owes whatever he can spare to his audience, and a sense in the back of his mind that YouTube, what with its fickle algorithm, is precarious. He made it this far by showing up, so he's mimicking a 20th century videogame designer to barrel forward with as much drive and Diet Pepsi as it takes.

You can see the effects in the resulting work—his Ultima series alone has only grown more detailed and elaborate with each new video.

Take The Tyranny of Virtue, Luke's retrospective on Ultima 5. By this point, the games have grown into a sprawling story about good and evil, acknowledging the player character directly as the Avatar; Ultima 4 had a remarkably robust morality system that rewarded virtuous in-character behavior, and Ultima 5 flips that on its head, using the previous game's virtues as the basis for a tyrannical villain's absolutist dogma.

It may have been easy to write these narratives off as cool for the time but pedestrian now, but Luke was up to the task of combing through the thematic implications thoroughly: he honors the game as a work of fantasy fiction in its own right, comparing its findings about the meaning of virtue to inspirations like Lord of the Rings and praising its willingness to interrogate its own messaging and encourage players to do the same.

Luke concedes that the narrative itself is a bit pedestrian in a post-Baldur's Gate 3 world, but the way the story leveraged the medium was wildly forward-thinking for its time. Those moments where the player's actions are brought into the story and roleplay is rewarded are, to Luke, some of "the most clever implementations in an RPG that I've ever seen to this day."

"The idea at the time of, like, 'you beat the game by acting virtuously,' would have been so abstract and such a hard sell for a lot of people," he said. "The narrative really reaches beyond the computer screen and becomes meta in a way that I would describe, as much as the word gets thrown around very flippantly these days, as genius."

His life somehow now revolves around talking about games in this way.

"Even on those 14-hour editing days, I'm glad that, at least for now, I don't have to get out into the car at -40 degrees Celsius to go lift drywall all day. That's a pretty positive change I would say," he laughed. "I know it won't last forever, so I'm always trying to make hay while the sun shines, and a big part of that is trying to improve my craft."

Through the ages

Onto new worlds of adventure

The Ultima retrospective is nearing its end. Luke plans to get through the main series ending with Ultima 9 and, beyond that, at least cover Ultima Online, which even taken alone was hugely influential in the world of MMORPGs. After that, there are a few "lesser games" to chew through: assorted cash grabs, phone games, and a glitchy mess PC Gamer's Rick Lane gave a 25. But while Luke's channel has always covered a variety of roleplaying games and accounted for a contingent of his audience with no interest in the games, ending a retrospective odyssey longer than a Ken Burns documentary will prove bittersweet.

"Ultima is not for everybody, but it's for me, because when I pull the brakes on it and when I give myself a chance to kind of recalibrate, by the time I'm diving into the next game, I can't wait to get back to the research component and see how things are changing in Britannia," he said.

"I've thought about Ultima pretty much every day for the last—I don't even remember when I started this series—two-and-a-half years ago at this point? It's been a huge part of my life, so I'm probably always going to bring it up. It's always going to be part of the identity of the channel, it's always going to be something that interests me, even just talking about it. What it did: how it changed the world of RPGs."

Score another one for YouTube: Despite what sometimes feels like an ongoing active effort to make it unpleasant to use, the platform that helped make a star out of guys like Pewdiepie, Dr. Disrespect, and Mr. Beast has made a big move into the mainstream media ecosystem with today's announcement that it will soon be the exclusive home for the Academy Awards.

That's right, beginning in 2029 the Oscars, "including red carpet coverage, behind-the-scenes content, Governors Ball access, and more," will be available exclusively on YouTube.

"The partnership also will include worldwide access for film fans to other Academy events and programs exclusively on the Oscars YouTube channel," the Academy of Motion Picture Arts and Sciences said. "This will include the Governors Awards, the Oscars Nominations Announcement, the Oscars Nominees Luncheon, the Student Academy Awards, the Scientific and Technical Awards, Academy member and filmmaker interviews, film education programs, podcasts, and more.

"In addition, through this holistic partnership, the Google Arts & Culture initiative will help provide digital access to select Academy Museum exhibitions and programs and help to digitize components of the Academy Collection—the largest film-related collection in the world, with more than 52 million items. It will be a true hub for film fans and will be accessible from around the world."

The Academy Collection is indeed a notable archive of film history, containing everything from Oscar-nominated films to manuscripts, film-related books and periodicals, photos, posters, manuscripts, and more. As far as long-term impact, digitizing those materials might be the most significant outcome of this partnership.

YouTube CEO Neal Mohan said in February that the online video platform had become "the new television," and he seemed correct—in the sense that it's riddled with inescapable ads that everybody hates (but that nonetheless make a ton of money) and appears increasingly concerned with ensuring that nobody under 18 is exposed to anything edgier than old episodes of Three's Company.

Setting aside the exaggeration-for-effect, this deal with the Academy is presumably closer to what Mohan actually meant: That YouTube is no longer the domain of computer nerds looking for old Tech TV clips, but is now running with the big dogs, like ABC—the current-but-soon-former home of the Oscars.

"We are thrilled to enter into a multifaceted global partnership with YouTube to be the future home of the Oscars and our year-round Academy programming," Academy CEO Bill Kramer and Academy President Lynette Howell Taylor said in a joint statement. "The Academy is an international organization, and this partnership will allow us to expand access to the work of the Academy to the largest worldwide audience possible—which will be beneficial for our Academy members and the film community.

"This collaboration will leverage YouTube’s vast reach and infuse the Oscars and other Academy programming with innovative opportunities for engagement while honoring our legacy. We will be able to celebrate cinema, inspire new generations of filmmakers and provide access to our film history on an unprecedented global scale."

It's interesting to see how far YouTube has come in the 20 years since it was launched, but I do have to wonder if there's not a certain risk involved in this—for the Academy, to be clear. I feel like the Oscars are of particular interest to an older demographic who might be less inclined to watch on an unfamiliar or inconvenient platform. Mohan said in February that "TV is now the primary device for YouTube viewing in the US," and I'm sure number-crunchers at both companies have taken pains to ensure the risk to audience numbers is minimal. But I'm also quite certain that I'm going to be hearing angry words about this from my mother, who is already plenty annoyed at having to (partially) figure out Netflix. The odds of her watching the Oscars on YouTube are roughly equal to the odds of her winning one.

For his part, Mohan called the Oscars "one of our essential cultural institutions," and seemed confident that even if he's leaving my mom behind, the opportunity to reach a new audience will be worth it: "Partnering with the Academy to bring this celebration of art and entertainment to viewers all over the world will inspire a new generation of creativity and film lovers while staying true to the Oscars’ storied legacy."

Felix Kjellberg, AKA PewDiePie, has shared in the Kjellberg Mail newsletter that after 13 years of being a gamer on YouTube, he is officially done with videogames for the time being. For those who have been following his channel for a while though, this won't come as much of a surprise.

Eight years ago, a video titled "Why I don't play video games anymore" was uploaded to the PewDiePie channel, explaining the reasons behind the lack of Let's Play videos. Since then, his channel has adapted into more lifestyle blogging and the occasional hobby vlog. Of course, there are a few games sprinkled in, with the most recent video featuring Geoguessr. But aside from that, it's not the content you would've expected to see ten years ago.

There's a good reason for this though, which PewDiePie explains in the November Kjellberg Mail newsletter. He shared, "Since becoming a dad, everything does get evaluated differently. Every hour has weight now in a way it didn't before. One of the biggest things I decided to drop was videogames. But here's the thing: if I really wanted to, there'd be plenty of time for me to play videogames. I've just decided that, if time is limited, I'd rather be doing something else."

The post continues to explain how he still plays 30 minutes or so of videogames every now and then, but claims that the "days of logging dozens of hours in one game are over." This isn't just due to being a father now either. Felix continues to say how it "ties directly into my YouTube channel. Since I've been so preoccupied with YouTube and gaming, I haven't actually done much to build myself and grow. So right now, I'm in a different mindset where I want to learn."

But just because the days of endless Let's Plays have come to an end, it doesn't mean the PewDiePie channel has come to a close. In fact, dedicating more time to different skills has inspired PewDiePie to post a wider variety of content which can be seen on his channel overview. For example, a video about his first PC build pulled in over 4.8 million views, while a follow up video titled "Accidentally Build a Nuclear Supercomputer" received 2.7 million.

These videos don't always earn a 100% positive reception, but that doesn't seem likely to stop this new era of content from being taking priority over any gaming videos. PewDiePie concludes his section of the newsletter with: "It's so funny to me when I release a video where I share whatever I've worked on. Some people get dumbfounded by how I can learn in such a short time. Some people call it out like 'Well, if I had infinite time like him I would too.' But the reality is I don't have infinite time. I just chose to spend it on what I think is more valuable, and seeing how people react like that validates that thought."

I must admit, I never really kept up with the still hugely popular YouTuber PewDiePie (to the tune of 110 million subscribers). So imagine my surprise when I discovered that these days he's doing bizarre hardware experiments, like bifurcating his PCIe slots to install a ludicrous amount of Nvidia RTX 4000 Ada GPUs, with an additional RTX 4090 thrown in for good luck. All for the sake of science and AI.

So, what does one do with a $20,000 rig (that hiss you're hearing is all the air leaving my lungs at once)? Well, in the beginning, the YouTuber—whose real name is Felix Kjellberg— ran protein folding simulations at night in aid of cancer research. Neat! But it wasn't long before Kjellberg disappeared into the rabbit warren of locally hosting massive AI models, eventually creating an AI-powered council and appointing himself as dictator.

Kjellberg explains that instead of asking Google or querying ChatGPT things like, 'who would be the best creator to collaborate with next,' he instead consults his AI council. He elaborates, "I have eight GPUs that can run the same model [specifically a version of Qwen]. Each one of them is a different council member with different personalities. So, it's a democratic process. I consult my council. They all give me an answer and then they vote. Democracy!"

Naturally, this system quickly fell apart. Kjellberg continues, "Only a couple council members were actually useful. The rest were just garbage. No one ever voted for them. Trash. So of course I had to kill them and replace them with new ones."

Kjellberg then made the mistake of making sure his 'AI council' had some idea that losing too many votes would cause their SQL database to be wiped from the face of existence. He explains, "The worst part [was] they colluded against me. They started voting strategically—helping each other even!"

But the YouTuber didn't stop there, realising he didn't need to arbitrarily limit himself to one AI model per GPU. He ultimately chose to use The Swarm, a madding crowd of 64 AI models which causes Kjellberg's web UI to lag when they all sound off. Though this experience has inspired the YouTuber to noodle around with his own AI, that doesn't necessarily mean Kjellberg is now a full-throated, evangelising AI tech bro.

For one thing, Kjellberg recently deGoogled his life amid privacy concerns, and revisits that theme here, raising the point of how hard it is to fully delete your data from ChatGPT. With his local setup, Kjellberg at the very least directly controls what the data The Swarm is pulling from (but admits giving it much of the data Google had previously collected on him).

For another, Kjellberg says he is not at all interested in AI video or image generation, saying, "I genuinely think AI art looks ass. It looks bad. You can tell immediately it's AI generated. And the art that does look cool has been straight up ripped off from another artist. So, why not just look up the artist?"

Though I'm hardly a PewDiePie fan (to say nothing of how I can only dream of getting within sneezing distance of a $20,000 rig), we can at least agree that human artists are just better.

When it comes to my e-waste themed sleep paralysis demons, none are more terrifying than the single-use, disposable vape. Why, oh why do these wretched things exist? And why, oh why do so many of them contain lithium-ion batteries that are in fact rechargeable? I'm breaking out in a cold sweat just thinking about it.

If only this scourge could be repurposed into something more useful than, say, yet another Doom box in every sense. YouTuber Chris Doel has apparently overheard my environmentally themed wailing, having already created a number of projects that repurpose these power cells into fast-charging power banks and e-bike batteries. His latest build dials "it up a couple of notches," creating something that not only generates enough energy to satisfy his power-hungry editing rig, but also powers his whole house using 500 disposable vapes.

In between repeated pleas to absolutely not try this at home, charmingly creative problem-solving is out in full force throughout. For instance, when a lithium cell discharges below three volts, it can't really be recharged. So, how do you sift through hundreds of salvaged vapes, sorting the still usable cells from the dead ones, in a timely fashion? Apparently, by deploying a pump from an old CPAP to puff on the vapes, thereby identifying which cells can still hold a charge without wasting hours dismantling already dead devices.

Even with this makeshift vaping device expediting some of the work, the project grew into a months-long endeavour. "It takes about two to three minutes to extract and desolder each cell. So, multiply that by at least 600," Doel reflects in the video, "And then it takes about two hours for each batch of cells to then be balanced, charged, and tested."

With any recycling effort being this labour-intensive, it's not hard to see why it became illegal to sell or supply single-use vapes in the UK earlier this year. But as Doel notes, that doesn't necessarily solve the environmental issue if retailers only supply the full refillable vapes and don't keep a similar amount of the vape refills on hand.

After hours of 3D printing and soldering, the hundreds of cells are mounted into a hefty powdered aluminium frame. To stop the whole thing from exploding, Doel installs a battery management system that he previously used in a 50 miles per hour mobility scooter build. On its own, the battery pack generates only 50 volts DC, so Doel then sets up an inverter to convert the current into 240 volts AC—before disconnecting both his home and workshop from the main electricity grid.

Not only does the disposable vape battery pack work, but the YouTuber is already considering future plans to take at least his workshop off-grid entirely. In the meantime, he charges the massive battery pack at night from the main electrical grid, making the most of off-peak pricing and proving that salvaged vapes can be more useful than Battlefield 6's smoke mortars.

There's no shortage of tools to get around Windows 11's installation nonsense, and that's for good reason. Some PCs can't even run the OS without tweaking, and potential users may want to avoid having to make a Microsoft account just to use their own rig. Recent reports suggest YouTube might be cracking down on content encouraging these workarounds, though.

YouTuber CyberCPU Tech has recently shared that they've had two videos removed in two days (via Tom's Hardware). Both of those videos are reportedly about ways to bypass Windows 11 requirements. One was a tutorial on how to install Windows 11 without a Microsoft account. The other was on how to install the latest version of Windows on unsupported hardware.

CyberCPU Tech says that both were taken down due to automated systems, but both were denied appeals. The first appeal was denied after just 45 minutes, and the second after just five.

In a previous video, he said that he had spoken to a support specialist (whom he believes could be AI) who said the Windows 11 videos promote "behaviour that encourages dangerous or illegal activities that risk serious physical harm or death." Naturally, we have to take the reported words of a support specialist with a grain of salt, and YouTube is an entity that doesn't always reveal specific reasons for its decisions.

The second appeal resulted in the following response: "Again, the warning strike you received was issued based on a violation of Harmful or Dangerous Content which prohibits content that encourages or promotes behavior that encourages dangerous or illegal activities that risk serious physical harm or death"

YouTube's appeals system is shrouded to some extent, with very few understanding the exact methodology. That being said, in many cases, a human review is part of the process. As answered in Google's YouTube Help section, "When our systems have a high degree of confidence that content is violative, they may make an automated decision. However, in the majority of cases, our automated systems will simply flag content to a trained human reviewer for evaluation before any action is taken."

We have reached out to YouTube to try and clarify what was taken down and why.

CyberCPU Tech says, "I had stated before that I didn't think Microsoft had anything to do with it. I don't believe that anymore. In fact, I believe they are entirely responsible for this." Despite this, there are still videos on YouTube showing how to get around hardware requirements, so whatever is going wrong here doesn't appear to be a blanket rule on the video-sharing platform.

One of the main issues here appears to be how tech YouTubers are to know which videos are safe to make. In the video following this, CyberCPU says, "I just don't know how tech channels on YouTube can survive when we have no idea what topics we can make videos on."

If you are familiar with YouTube, you will likely spot this as a bit of a trend. Many creators self-censor parts of their videos they believe could cause them problems or limit reach, and YouTube isn't always upfront about the specific elements that cause problems.

As such, we only have guesswork and CyberCPU Tech's word to go on here, but it's still a worrying trend to see video tutorials like these taken down, even if it's purely being done by YouTube's AI bots.

As always, I'd encourage checking out Rufus or working around Microsoft account requirements, should that be how you want to run your rig. After all, there are still plenty of tools widely available to customise your PC, even if Microsoft has cracked down on some of them this year.

YouTube's lengthy policy on violent content is getting a little longer, and Grand Theft Auto videos in particular may trigger enforcement of a new rule related to depictions of torture and mass violence.

"In addition to our existing guidelines around graphic gaming content, we will age-restrict an additional small subset of video game content featuring realistic human characters that focuses on scenes of torture or scenes of mass violence against non-combatants," reads an update posted this week.

The video platform will make its judgment based on three criteria: The duration of the violent scene or scenes, whether or not the violence is the focus of the video, and whether or not the characters in the game are realistic humans.

Unless the Postal series sees a huge resurgence in popularity, GTA 5 and the upcoming GTA 6 are the most likely games to put the "mass violence against non-combatants" guideline to the test, though from the sounds of it only videos that make mass murder a particular focus will be hit with an age restriction under this rule. Clipping a virtual pedestrian with a car during a mission might be fine, but making a video about driving through crowds could be flagged.

YouTube is also expanding its definition of gambling content to include the gambling of videogame skins and NFTs, and will now age gate "content that depicts, promotes, or facilitates social casino sites."

Videos which YouTube age gates can only be viewed by logged-in users who are 18 or older.

The new rules take effect November 17, and may be applied retroactively, causing existing videos to be age gated or removed, but any such actions "will not result in a strike," says YouTube. More on the policy update can be found on the YouTube support site.

YouTube has joined the ignominious list of tech companies rushing to settle lawsuits brought by US president Donald Trump. The Alphabet-owned platform has agreed to pay $24.5 million to settle a lawsuit filed after it suspended Trump's YouTube account following the riot at the US Capitol on January 6, 2021.

Trump has accused various tech companies including YouTube of political bias and censoring conservative voices, and both Twitter/X and Meta have already settled lawsuits relating to the suspension of his accounts. In January 2025 Meta agreed to a $25 million settlement, with $22 million of that going to Trump's presidential library, and in February X settled for a reported $10 million.

At the time of the Capitol riots, these firms had said that Trump's posts risked inflaming tensions and inciting further violence. All of Trump's accounts on these platforms have been reinstated.

Needless to say, Trump took to social media to celebrate the win, posting an AI-generated image on X of what appears to be a distressed-looking Neal Mohan, CEO of YouTube, handing over a giant cheque for $24.5 million to the president, who is grinning and giving a thumbs up. The cheque reads: "Settlement for Wrongful Suspension."

On Trump's Truth Social platform the same picture comes with a wall of triumphant text, featuring the president's typical fondness for caps lock:

"YouTube SURRENDERS! Pays President Trump $24.5 MILLION for illegal ban! This MASSIVE victory proves Big Tech censorship has consequences. Every shadowbanned patriot deserves justice! Trump fought for free speech and WON! Repost if ALL banned conservatives should be paid!"

All very normal. YouTube has been feeling the heat from the Trump administration more generally, and has not exactly covered itself in glory. It and other social media firms have agreed to take steps to loosen content moderation on their platforms, and last week YouTube said it intended to restore accounts that had been banned for posting misinformation about the 2020 presidential election and Covid, among other topics.

"YouTube values conservative voices on its platform and recognizes that these creators have extensive reach and play an important role in civic discourse," says a YouTube statement sent to a Republican-controlled congressional committee.

Under the settlement YouTube does not admit any wrongdoing, nor agree to make any policy changes. The $24.5 million will see $22 million going to the Trust for the National Mall, a group aiming to raise $200 million for the construction of a new White House ballroom. The remaining $2.5 million will be paid to other plaintiffs in the case, including the American Conservative Union.

I think we can all agree that dealing with cheats is one of the worst parts of multiplayer gaming. Developers continue to add anti-cheat software to their games, cheating software continues to become sneakier and sneakier to get around it, and the whole cycle spins around again. One YouTuber, however, has created a device so ingenious I think we're all doomed if it ever takes off—a virtually undetectable mouse pad-moving aimbot.

Master of mischief Kamal Carter has put an astonishing amount of effort into their latest creation, a devilish cheating device with four simple goals (via Hackaday). One, it needed to be better than them at popular hero shooter Valorant. Two, it also had to be better than their friends. Three, it had to deliver a level of performance comparable to a Valorant esports pro, and four, you guessed it, it needed to be undetectable in-game.

Carter first manually trained an instance of YOLO, an AI object detection algorithm that works by detecting bounding boxes and identifying class probabilities, in order to detect enemies in Valorant's shooting range challenge. They then classified a hitbox area for the aimbot to shoot at for each defined target, aiming for the center mass by taking advantage of Valorant's similarly-sized character models.

For the mechanics of the hardware itself, Carter went through multiple iterations of potential mouse-moving devices, before landing on a DC motor-powered, CNC-constructed frame that physically tilts a plywood board, on top of which rests a modified optical gaming mouse inside a thin plastic holder to keep the squeaker from moving away from the center of the pad. The mouse is "clicked" with a switch bypass relay, toggled by the bot.

The software-linked motors respond to onscreen targets via voltage adjustments, speeding up and slowing down their responses based on how far away the targets are onscreen. Carter admits this method causes a small amount of "mechanical chaos" as the bot picks its targets, but ultimately decided it was a good thing, as it mimics the overshoot and corrections made by a real human player—making the bot less detectable than it would be otherwise.

The response time of the mouse pad-moving motors isn't as instantaneous as Carter would like, but it appears to be good enough (when linked to the trained AI model) to beat their eight out of 30 target score in Hard mode, where each target remains onscreen for a mere half a second. Pro players can score anywhere between 22 and 30 successful hits at this setting, whereas Carter's bot managed a top score of 26 out of 30 after a day's worth of fine-tuning.

Carter admits the bot could be tweaked to become even better, hitting a maximum 30 out of 30 score, but the fact that the results aren't perfect again serves their ultimate goal of creating an aimbot that responds like a human player. Even the top pros don't always manage perfect scores, so a "solid, consistent performance" makes the bot less likely to be detected by the game's anti-cheat software.

The good news is, the aimbot only currently works with Valorant's shooting range challenge, and hasn't yet been integrated with something that could move the player character around in a genuine match. It's a shooting gallery master, and I reckon that's where it should probably remain for now.

So, top points for creativity, lateral thought, and impressive engineering. Minus a million points, however, for creating what might be the start of undetectable aimbots to come. My aim is already suffering after years of… well, being alive, so you don't need a tiltable table to beat me in a multiplayer match. Shout "boo" over voice chat while I'm lining up a sniper rifle shot, or something. That'll probably do it.

This may scandalise some younger readers, but once upon a time you'd turn on the telly and simply watch what was on. The question of what to watch these days has only gotten more complex, with a wealth of streaming services now vying for your eyeballs. Well, Amazon may be touting AI as the answer—I mean, just judging by the undisclosed amount they've thrown behind startup Fable.

The Edward Saatchi-owned outfit has already been around for a few years, debuting with VR game Wolves in the Walls back in 2019. What's piqued Amazon's interest more recently is Fable's AI-generated 'streaming' service Showrunner (via Variety). Hop into the Discord, and you can start mashing out prompt-based animated scenes that are all about as bleak, monotone, and ugly as I was personally expecting.

Having already run as a closed alpha test with 10,000 users, Showrunner is currently free to use. The recent public launch offers a few 'story worlds' to plug prompts into, all powered by Fable’s SHOW-2 AI model. Exit Valley is a popular 'show' on the Discord, but very much reads to me like shockingly average adult animation that's trying way too hard. Using Exit Valley's AI stage you too can generate scenes like this one, where a cartoony Tom Cruise jabbers with a Kim Kardashian that refuses to skip arm day—and also inexplicably sounds progressively more Australian with every rep.

Fable wants to eventually implement a 'credits' system for creators, planning to charge between $10 and $20 a month to generate more AI scenes. However, all Showrunner content will remain free to view in the hopes of encouraging folks to watch and share clips on third-party video platforms like YouTube (where Showrunner's f-bomb heavy schtick may not even get demonetised).

Edward Saatchi told Variety that the vision for Showrunner is to create something "playable," in a not entirely dissimilar way to mashing dolls together in a dollhouse. He said, "The ‘Toy Story of AI’ isn’t just going to be a cheap ‘Toy Story.’ Our idea is that ‘Toy Story of AI’ would be playable, with millions of new scenes, all owned by Disney."

Dropping the big D isn't just a powerful name drop, as Fable is apparently in talks with the house of mouse about licensing its IP for Showrunner. But even with such upfront ambition, even Saatchi is saying: "Maybe nobody wants this and it won’t work."

In conversation with Variety, he says this within the wider context of his earlier ventures into VR not taking off as he had hoped. But Saatchi also concedes that "today AI can’t sustain a story beyond one episode," and that the technology is perhaps better suited to "deeply episodic shows with characters largely resetting every episode — sitcoms, police procedurals, space exploration."

I mean, points for self-awareness, but I'd argue even traditional intrigue-of-the-week shows often feature season spanning narrative arcs or even simple callbacks. Don't worry, I'll spare you my impassioned defence of Elementary as the genuinely better take on 'Sherlock Holmes, but make it a cop procedural' from the 2010s. Anyway, besides that, I was already looking at Showrunner's dollhouse and loudly wondering what the point of it all is.

For one thing, deciding what I want to watch via traditional streaming services already requires work amounting to a low-level research project—I already enjoy that work, so why would I put my energy into the well-draining black hole of AI? For another thing, 'playing' with established characters and IP just makes me think about the good old, wonky days of machinima—to say absolutely nothing about the fanfic burning a hole on my hard drive.

With these creative outlets and more already available to me, I don't really feel the need to play within the confines of Showrunner's corporate, AI playground.

YouTube warned us this was coming back in February. Now it's actually rolling out a new AI feature to determine the age of viewers and then feed them age-appropriate content.

According to an official YouTube blog post, this machine-learning approach to content filtering and user account classification has already been running in some unidentified markets for an unspecified period. But over the next few weeks, YouTube will "begin to roll out machine learning to a small set of users in the US to estimate their age, so that teens are treated as teens and adults as adults."

YouTube says it will closely monitor the technology, "before we roll it out more widely." The video streamer explains that the system allows it to "infer a user’s age and then use that signal, regardless of the birthday in the account, to deliver our age-appropriate product experiences and protections."

If YouTube, or rather YouTube's bot, decides a teenager is watching, a number of measures can come into play, including disabling personalised advertising, turning on digital wellbeing tools and adding safeguards to recommendations. This could also include limiting repetitive views of some kinds of content. Exactly what these tools and measures amount to, such as "digital wellbeing", isn't entirely clear.

As for how the AI system makes that judgement, YouTube says it interprets, "a variety of signals that help us to determine whether a user is over or under 18. These signals include the types of videos a user is searching for, the categories of videos they have watched, or the longevity of the account."

But what if YouTube gets it wrong? "If the system incorrectly estimates a user to be under 18, they will have the option to verify that they are 18 or over, such as using a credit card or a government ID. We will only allow users who have been inferred or verified as over 18 to view age-restricted content that may be inappropriate for younger users," YouTube says.

It will be interesting to see how accurate the AI age verification system proves to be. Certainly, AI is posing YouTube, or at least its viewers, at least as many problems as it is solving.

YouTube's earlier February blog post, which first mentioned the use of machine learning to filter user accounts, leaned more broadly into the benefits of AI, highlighting tools for creators, including AI-powered auto-dubbing and AI-generated video backgrounds and soundtracks.

On the other hand, a huge quantity of AI-generated slop is now uploaded to YouTube daily, and the company is currently either unwilling or unable to filter it all out. Overall, it remains to be seen if AI will more broadly prove a saviour or a curse for the platform. Oh, and us, the people who actually use it.

Back in 2022, YouTube introduced new rules for creators on its platform that demonetized videos containing "strong profanity" profanity in the first seven seconds. The results were not great: Unclear policies and uneven enforcement left more than a few YouTubers up in arms over the potential threat to their livelihoods.

Several months later, YouTube eased up a little bit, saying videos containing "moderate profanity" would be eligible for full monetization while those using "stronger profanity" in the first seven seconds or "repeatedly throughout the majority of the video" would be eligible to run "limited ads." Better, yes, but by much, and still unnecessarily vague and confusing.

Now it's finally quit horsing around: As of today, F-bombs and all the other good words will no longer result in demonetization, even if they're dropped within the first seven seconds.

The previous policy was put into place, Connor of YouTube explains in today's video, because advertisers wanted "distance" between their ads and the swears. That apparently is no longer the case: "Those expectations have changed and advertisers already have the ability to target content to their desired level of profanity."

So what kind of profanity are we talking about here, exactly? "Moderate profanity is words like 'asshole' or 'bitch'," Connor says. "Strong profanity includes words like 'fuck.' I don't think we need to use more examples, but you get the idea." I do get the idea, but it'd still be more fun to have more examples cited.

Connor also points out that this is not a green light to turn every YouTube video you make into a Trailer Park Boys compilation.

"You have to pick and choose your fucks carefully," he says, warning that a "high frequency" of strong profanity will still violate YouTube's advertiser-friendly guidelines. Monetization will also be limited for videos that use either moderate or strong profanity in titles or thumbnails, and the use of "extreme profanity"—racist or homophobic slurs, specifically—will still result in complete demonetization.

It's not a complete free-for-all then, but it still strikes me as a meaningful improvement over the previous rules, which were honestly kind of silly: Who decided that seven seconds was the appropriate window between the end of the ad roll and the first f-sharp? More importantly, though, is the practical upside. Eliminating that seven-second swear buffer means it should be harder for YouTubers to accidentally break the rules now, and for those who rely on their channels for income, that's a pretty big deal.

As is the case with all good things, the first time I heard the 'leak' of Tyler, The Creator's 'Don't Tap the Glass' was on TikTok. A 'club banger' in the style of Avicii or even The Chainsmokers, it's not only noticeably shallow, but very much unlike anything else by Tyler, The Creator. That's because it's not. Instead, it's part of a breadth of YouTube videos marrying the clickbait stylings of old-school YouTube with the newfangled tools of generative AI. What a world we live in.

Unlike when a creator would say GTA V had leaked in 2010, only to show nothing to back up the claim, what makes this particular use of generative AI so effective is that it can pass the sniff test for some internet users.

Posted by user KLODJAN, the video titled 'Tyler, The Creator - Don't Tap The Glass! (Music Video)' launched just before the release of Tyler, The Creator's new EP and now has over 200,000 views.

This same user has been using this tactic for months now and has seen some pretty big results. A 54-second video titled 'Sabrina Carpenter - Manchild (Music Video)' cashed in on the hype of Sabrina Carpenter's song of the same name and has since amassed over 100,000 views.

What makes this specific sound unique is its context collapse. It has found its way onto TikTok, where over 800 posts use it, with another 50+ posts using a different upload of the sound. As well as this, videos with over 100k likes are making fun of it. Like the Sabrina Carpenter AI music video, it's a flash in the pan, but one that is getting some serious views.

YouTube put out a statement in 2023 on the use of AI in the music industry. In it, YouTube said it "will embrace it responsibly together with our music partners." It claims it will support creators who use generative AI but will "empower creativity in a way that enhances our joint pursuit of responsible innovation." These rather nebulous claims leave a lot of room for interpretation.

Clicking on KLODJAN's channel, the videos are currently monetised. YouTube does pay creators for AI-generated videos (as long as they're provably unique), but these videos being low quality, based on false premises, and going live without AI disclosures does suggest it won't stay monetised for very long.

However, even if these channels lose monetisation, that doesn't mean YouTube won't put ads on the site; it just means creators are no longer incentivised with money to keep making them. Views may not always be monetisable, but subscribers are still a valuable asset for any channel. This is part of a broader trend happening on YouTube right now. Just looking for trailers for new games (like Elder Scrolls VI) or movies (like Avengers: Doomsday) shows a gulf of AI-generated trailers hogging up the search function.

I didn't anticipate having to debunk my YouTube feed anytime I search for something new now, but alas, the 'wonders' of generative AI have made it so.

An Italian YouTuber who specializes in retro gaming says he's facing criminal charges and a possible jail sentence as a result of reviewing retro gaming consoles on his channel, some of which came with pre-loaded ROMs.

Francesco Salicini, who streams as Once Were Nerd, said in a video posted on July 14 (via Android Authority) that the ordeal began on April 15 when Italy's Guardia di Finanza, a law enforcement agency that focuses on financial crimes, showed up at his mother's house with a search warrant.

Believing he'd done nothing wrong, Salicini opted to fully cooperate on the spot: He invited the police to his studio, where they seized more than 30 consoles, along with his mobile phone, which they held for more than a month.

Salicini said in his machine-translated video that he's accused of violating article 171 of Italy's copyright law, which he believes is because some of the consoles he reviews on his channel ship with microSD cards filled with games. In a GoFundMe page (Google translated) raising money for his legal defense, Salicini said the problem appears to be his "sponsorship" of those consoles—presumably a reference to sponsored promotional reviews.

"While I've always insisted that these consoles can only be used if you have, as I do, the original copy of the game, and that they're also available for sale on Amazon, eBay, and various other stores, I've never 'sponsored' them, but rather 'reviewed' them with the utmost objectivity, citing both the pros and cons of each device," Salicini wrote.

Salicini isn't entirely certain of what he did to violate the law, he explained, because police aren't required to inform him of specific charges until after the investigation is completed. But the potential penalties are stiff: Article 171 of Italy's copyright law allows for prison terms of between six months and three years, however, and—because the statute was written in 1941—a fine of up to 30 million lira, which works out to a little over €15,000.

For now, Salicini's YouTube channel, with more than 220 videos, remains available; he said in his GoFundMe that the Guardia di Finanza has "blocked" his social media channels, although his Instagram and Discord are still up. While it's possible the charges against him will ultimately be dismissed after the investigation is concluded, he said in his GoFundMe that he expects it will ultimately go to trial—and yes, he now has a lawyer. I've reached out to ask for comment and will update if I receive a reply.

At the start of this year, after YouTube inexplicably decided that I was not, in fact, a proud resident of Argentina, it shunted me off my alarmingly cheap YouTube Premium package back into its regular, un-premium offering. It's terribly disruptive, which I imagine is the point: ads before I start a video, during a video, next to video previews, as far as the eye can see. No, I don't want to use Grammarly, thank you.

So imagine my delight on learning, via The Verge, that YouTube's top brains have cracked how to make the whole experience even worse for the end user, thanks to the infinite power of Google's Gemini AI. At this week's Brandcast event (a big networking and presentation carnival for the platform's advertisers), YouTube gleefully announced "Peak Points." That's its name for a new system that lets Gemini identify the "most meaningful, or 'peak,' moments within YouTube’s popular content."

So far, so acceptable (though I don't know how that's markedly different from YouTube's already-existing "most replayed" feature), but then comes the sucker punch—advertisers can take advantage of Gemini's Peak Points feature to "place your brand where audiences are the most engaged".

In other words, YouTube is turning AI to the task of identifying "contextually relevant" parts of videos on its platform to cram ads into, ensuring that they're carefully placed to hit viewers when they're "most engaged" with the content on their screen. I don't know about you, but an ad suddenly Kool-Aid Manning through the wall when I'm most engaged with the thing I'm watching feels like a recipe to reach new peaks of frustration. I dunno if that's really the reaction advertisers want.

But, frankly, I wonder if it's even about any of that, so much as it's about giving Google an opportunity to say 'AI' on a stage and trigger the Pavlovian response that has the world's executive class handing over wads and wads of cash. Call me a terribly cynical Luddite (and, hey, the Luddites had a point, thank you), but this feels like another mark in the ledger for AI being a technology that has investors and capitalists incredibly excited, but that offers very little to humble proles like you and me except a worse, more frustrating world.

YouTube is finally doing something about all those AI trailer videos that have been cropping up across the platform. If you've been on the internet in the past few years you've likely seen one of these AI film trailers. Maybe you've even left a poorly generated AI comment on them.

They often feature cuts of famous actors spliced in from other projects, or as is becoming more common, straight up fake generated footage to advertise equally non-existent movies. This is all done in the name of revenue, and so YouTube has suspended several large AI trailer accounts from the company's paid partner program.

This all comes after Deadline began investigating the rise of AI trailer videos on YouTube. Since then, the website reports the suspension of two YouTube channels, Screen Culture and KH Studio, from earning revenue from their AI videos. The channels have 1.4M subscribers and 724,000 subs respectively, and that's just on their primary accounts.

"Our enforcement decisions, including suspensions from the YouTube partner program, apply to all channels that may be owned or operated by the impacted creator." reads a statement from YouTube, so it should include any of their smaller linked accounts too.

Taking monetisation away from these accounts is a good start, as it's clearly the driving goal. Not just for these accounts, but also from others that may be claiming their own revenue. Deadline's investigation also brought to light that prominent Hollywood studios, including Warner Bros. Discovery and Sony, have claimed ad revenue on Screen Culture trailers. So even the companies in control of the IP aren't particularly motivated to put a stop to these videos, especially while checks are still rolling in.

With the suspension on these major accounts, hopefully we'll see a huge dip in their propensity on YouTube. Without financial remuneration the creators of these videos should be less incentivised by spreading misinformation.

There's also the glimmer of light that suggests this could be the start of YouTube taking more serious action against AI content on the platform. Though I worry that's probably a little bit too much to hope for in one day.

What year is it? The ninja behind early 2000s internet video series Ask a Ninja clearly anticipated the reflexive reaction to his first new video in nine years, leading off with an instant parry. "What year isn't it?" he says within the first three seconds.

"If you're still believing in time, my friend, I don't know what to tell you. Have you ever heard of the internet? Everything lasts forever on the internet! Except MySpace photos, or if you had a Flickr Pro account. 2025, I'm still alive!"

I cannot believe that the Ask a Ninja guy looks and sounds exactly the same as he did 20 years ago, when pirates vs. ninjas and the general concept of bacon were the most popular topics on the internet. Ask a Ninja sprang forth from the innocent time when memes lasted for years instead of days and putting on a balaclava, doing a funny voice and uploading 360p videos could get you a book deal and lead to you interviewing Will Ferrell. Seeing him suddenly pop up and do new jokes in 2025 makes me feel incredibly old. That balaclava better be hiding some gray hair.

But I think it's also… making me happy? Maybe confused. Unsettled. Yesterday I was prepared to write the return of Ask a Ninja off as a freak incident, but maybe there's something in the water (or in the tubes), because today Homestar Runner published a new video celebrating its 25th year online. It's a song called "Back to a website," and appropriately celebrates a time when "the whole internet wasn't just on four websites on peoples' phones."

I have some deep, unhealed wounds from being in high school when Homestar Runner was considered peak comedy by 15-year-olds—I would really be okay going the rest of my life without hearing Strong Bad's voice—but this song still speaks to me. As PC Gamer's Joshua Wolens recently advocated, the internet was a richer place to be when it was made up of a million thriving pockets of weirdos with no monolithic social media corralling everyone into the same centralized space.

But I'm actually a little optimistic about what feels to me like a groundswell of interest in rebuilding some version of the 2000s internet with more personal blogs and decentralized social media that values people over platforms. In the last year I've seen more people adopt the concept of "POSSE," or Publish (on your) Own Site, Syndicate Everywhere; push back against Substack's attempt to own newsletters in favor of independent publishing; and go all-in on making the fragmented web easier to navigate.

Maybe the Homestar Runner video is simply a cheeky way to celebrate the site's longevity, but "let's go back to a website" manages to feel like quite a timely message to me in 2025. There are some good jokes in there about web rings, guest books and hit counters, quaint fossils of the early internet. Yet Homestarrunner.com still uses Flash courtesy of open source emulator Ruffle, and it's proof the video has a point. Clicking around that website is just fun in a way that most sites today are not.

Ask a Ninja's return video is timely in its own way. "Is the ninja community being affected by tariffs?" he's asked, which gets a smooth pivot to "Ninjas are way more affected by tear-offs—we have to pay an extra duty for every arm that we remove!" I'm not saying that the ninja returned from his long slumber because the state of the world was so bad it awoke him in grand Arthurian tradition just when we needed him most. But his second Return Era video, released three days ago, is titled "Zune vs iPod"—clearly this is a man who knows how to deploy the healing power of Remembering 2009.

I'm not sure two incidents are enough to declare a trend, but be on the lookout for the Potter Puppet Pals or the Peanut Butter Jelly Time banana to reappear at any moment. If this isn't proof enough the early 2000s internet is making a comeback, I don't know what is.

Longtime FromSoftware dataminer Zullie the Witch has uploaded a new video that interrogates one of the studio's most memorably strange non sequiturs: The giant guy made of rope who carries you from the kingdom of Ashina to the celestial Fountainhead Palace. It seems like there was once much more to Mr. Rope-a-Dope than we saw in the final game.

Much like the giant crow who takes us to Lordran in Dark Souls 1, or the multiple convenient transportation gargoyles in DS1 and 3, the giant rope guy in Sekiro always felt like a bit of a "well, fuck it" area transition.

Thankfully, the absurd and nonsensical works just fine with FromSoft's surreal, disconnected house style. I did learn through Zullie's video that rope man was likely based on the real-life Japanese Wara ningyō dolls, just at a massive scale.

Looking into the creature's associated files in Sekiro, Zullie uncovered unused AI scripts for the giant Wara ningyō which imply that it was once supposed to have active gameplay associated with it, rather than the single cutscene it has in the final game.

That, of course, begs the question of how such a massive creature would have even worked with Sekiro's gameplay. Zullie measured the rope man as standing 191 meters tall⁠—that's more than twice the size of the Statue of Liberty, and almost halfway to the height of the Sears Tower in Chicago.

One possibility pointed out by Zullie is that you didn't fight the rope man directly, but instead did battle with normal-sized enemies while climbing the thing. This is supported by Sekiro concept art of the rope man that shows various enemies hitching a ride.

That would have presented its own issues, though. Sekiro's jumping and free movement doesn't seem like a great fit for such a constrained play space, particularly if it was moving, as the AI scripts imply. Ultimately, I'm pretty happy with what we got: The sort of utterly surreal, unexplained oddity that FromSoft is so good at creating. You can follow Zullie the Witch on YouTube for more deep dives into FromSoft esoterica.

YouTube is conducting a "platform-wide review" of videos and channels on its platform to ensure age restrictions are being properly applied—and, for the record, says Balatro gameplay videos should not be rated 18+ unless they actually violate its gambling policy or other community guidelines.

The problem came up over the weekend when Balatro creator LocalThunk took to social media to say that Balatro videos "are being rated 18+ on YouTube now for gambling."

"Good thing we are protecting children from knowing what a 4 of a kind is and letting them watch CS case opening videos instead," LocalThunk wrote. And, hey, fair point: Videos of CS2 case openings, which have well-documented connections to actual gambling, are hardly in short supply on YouTube.

As silly as it is, it's also a genuine headache for YouTubers: Having an age restriction applied to a video limits who can see it (and thus who it can be recommended to via YouTube's algorithm) and can also impact monetization, as some advertisers won't allow their ads to run on age-restricted videos.

Ah, but it was all a mistake. "TY for flagging!" the TeamYouTube account posted on X in response to LocalThunk's message. "Videos featuring Balatro gameplay should not be age-restricted. We're reviewing the age-restrictions on these videos and conducting a platform-wide review. Thanks for your patience as we fix!"

A YouTube spokesperson confirmed the response was legitimate in a statement sent to PC Gamer: "Videos featuring Balatro gameplay should not be age-restricted. We're fixing the issue and conducting a platform-wide review."

YouTube suggested that Balatro fell victim to the recent changes to YouTube's policies on gambling-related content and how those policies are enforced, saying it welcomes feedback on its enforcement policies as it continues to fine-tune how they're applied. It also encouraged other content creators who are hit with an inappropriate age rating to file an appeal.

Those of you who've been following along will know that this isn't the first age-rating-gone-wrong rodeo for Balatro, a game that makes some superficial use of poker mechanics but is free from any in-game spending or external purchases.

In December 2024, PEGI—the Pan-European Gaming Information rating system—rated Balatro 18+, saying it teaches "knowledge and skill [that] could be transferred to a real-life game of poker." That rating was eventually walked back too, although it took a couple months to get that one cleared up.

Split keyboards are not for everyone. But if you fancy the hands-spread typing experience and don't feel like paying $200+ for something like the Kinesis Freestyle Edge, Japanese YouTube channel Nomoluk Report has shown a quick and easy method for cutting a regular one in half. Okay, I lied. Nothing about this looks quick and easy at all.

As spotted by Hackaday, Nomuluk Report has done the unthinkable and taken a rather-fancy-looking saw blade to a regular mechanical keyboard and chopped it in half, like an amateur magician having a very bad day.

After hacking the innocent keyboard into what looks like an ignominious death, the channel then gets to work with a soldering iron and wires the traces back together, creating a device that very much looks like it's begging to be put out of its misery. A quick bit of cable conduit wizardry later, and the device looks perfectly usable—although still begging for its electronic life.

My favourite part comes at 1:37, where the hacked-in-half keeb is demonstrated to only work on one side before it's wired back together. Who would have thought chopping a circuit board in half would ruin at least some of its functionality? Not me, I tell you what.

"I regret why I started doing this," the narrator says as they wire together the disparate halves. "The detailed work was too hard so I escaped and came to see the sunset."

Looking at that nightmare of wires, I'm really not surprised. Still, the end result looks surprisingly functional. Even the RGB lighting appears to work, although the spacebar is left dangling off of one side of the board.

The switch itself needed to remain on the right-hand side for a "clean" cut, so there it remains, a grim reminder of the intact keeb that once was.

Later in the video a split membrane keyboard is shown wired together with a similar method, although this poor peripheral didn't even receive conduit surgery to tidy up its trace wiring. If you're a keyboard enthusiast, look away now. This really isn't pretty.

If I was to describe my face while watching this video with a single word, I think I would choose "aghast." Anyway, it's done now, and both keyboards look to be fully functional. If, it must be said, not entirely happy with their new lives as two separate-yet-ever-connected peripherals.

There's hackery and then there's, well, hackery. Still, if you too have a desire for a split keyboard and happen to have an old model lying around… leave it alone. Keyboards have enough of a terrible existence if you ask me, taking a relentless bashing from our greasy fingertips and absorbing spillages, crumbs, and in my keebs case, dubious writing.

Leave this one to the, err, professionals. Or don't, if you've got some sort of grudge against regular keyboards that simply must be fulfilled. Now, if you'll excuse me, I'm feeling slightly woozy. I think I need to go for a lie down.

YouTube channel Hardware Busters claims it purchased what looked like an AMD Ryzen 7 9800X3D CPU from Amazon for processor testing purposes. Good for them! Upon opening the box, however, a few things stood out. One, it wasn't the correct socket and heatspreader design for an AM5 chip, and two, an eagle-eyed viewer noticed that the supposedly silkscreened logo on top appeared to be a sticker.

Closer inspection (and a bit of fingernail prying) revealed that underneath the sticker lay an AMD FX 4100, a 14-year-old Bulldozer chip that can be found on the used market for around $10.

The quad-core, four-thread FX 4100 is a socket AM3+ chip with a 3.8 GHz boost clock and support for DDR3 memory. The $479 AMD Ryzen 7 9800X3D, however, is an eight-core 16-thread, 3D V-Cache-imbued powerhouse of a gaming CPU on the AM5 socket with support for DDR5 memory, and a 5.2 GHz max boost clock.

Basically, they couldn't be further apart if one was one of the best CPUs for gaming, and the other was a ham sandwich.

So what tomfoolery has occurred here? Well, this is likely a case of some unscrupulous Amazon customer buying a near-$500 gaming CPU, then marking it as faulty and returning a much, much cheaper processor back to Amazon—with a sticker on top.

Returned electronic components are unlikely to be checked by someone qualified to tell whether they're the correct model or not. After all, it came back in the right box, and even removing it from the packaging reveals what looks like the correct CPU model to an untrained eye.

The unscrupulous buyer gets to keep their mega gaming CPU for free, a nearly-worthless processor is lost in exchange, then the box is plastic wrapped, put back on the shelves, and some poor soul buys it for their next gaming rig. Or in this case, test bench machine.

As Hardware Busters concludes, it's vitally important that you check hardware components inside the packaging as soon as you receive them to see if they really are the ones you paid for, even from an otherwise reputable retailer.

Return windows vary, and the sooner you can get a mislabelled (or in this case, poorly-stickered) component back for a refund, the better.

At least in this case the differences were obvious, but it might not be quite so clear in the case of, for example, a GPU. The less-tech savvy among us could quite easily fit a lower-model graphics card with misleading branding on it without realising they'd been ripped off, so always verify your hardware after installation with handy hardware diagnostic tools like CPU-Z or HWInfo.

The more you know, etc. What I want to know is, who's going to all the trouble to print fairly convincing stickers in order to rip off retailers for gaming CPUs? Still, it's more convincing than writing "Ultra fast AMD superchip" on the heatspreader in marker pen, I guess.

It can be hard to keep count of all of the online subscription fees I'm paying these days; a few quid here or there for films, music, and my terrible, no good Final Fantasy XIV: Online habit that I just won't quit. For the low, low price of a couple of bob a month, I get to live the dream of being a magical cat girl adventuring across Eorzea and beyond—which continues to make it worth my while. Unfortunately, YouTube now wants to add to the tidal wave of subscriptions by offering a cheaper Premium tier with fewer features.

For a long time I've resisted taking the plunge into yet another subscription with YouTube Premium for $13.99 a month, enduring 30-second ads (or even longer) when I lazily open the app on one of my consoles. Well, now there's a lower price point to tempt me, with a pilot YouTube Premium Lite tier rolling out across the US for $7.99 a month.

Naturally, you're getting less for, well, less. Just for a start, this Lite tier only offers "most videos ad-free," whereas the more expensive Premium tier offers the unqualified "ad-free" perk. Premium Lite subscribers will still get videos on "gaming, fashion, beauty, [and] news" ad-free, but they'll lose out on YouTube Music and the background play feature. That said, if you're already flinging money, say, Spotify's way, that may be a moot point for you.

YouTube recently announced that YouTube Music and Premium combined have 125 million subscribers across the world—so at the very least I know it's not just me who finds those lengthy ads deeply annoying. While the cheaper Premium Lite tier is new to the US, it's already available in a similar pilot scheme capacity in Australia, Germany, and Thailand.

Given that ads made YouTube $10.4 billion in the last three months of 2024 alone, doubling down on subscriptions is hardly unsurprising. Add to that YouTube's enduring popularity, with viewers apparently streaming 1 billion hours of content to their TVs on the daily, and it seems my only recourse is to not so quietly grumble about the platform's life-force draining ad spots. I'm just thankful that ads playing when you press pause seemingly didn't stick around.

YouTube claims that the Premium Lite tier will "create additional revenue opportunities for our creators and partners," though I can't help but feel a little sceptical. To be clear, I've no doubt YouTube and its ad partners will rake it in. However, I fear many creators will enjoy little benefit, if any—seeing the pennies similar ad-free subscription services like Spotify throw to even successful creators isn't exactly filling me with hope.

In a recent episode of Release Notes, Chief Product Officer Johanna Voolich explained the intention is not to frequently tack on tiers but to keep the overall subscription service "somewhat simple." Still, if YouTube Premium introduces a tier that excises AI-generated content—and kicks those reply suggestions to the curb—then perhaps I'll yet become a loyal subscriber.

As someone who writes for a living, having someone jumping into my Google Docs to make sure I haven't written something too abhorrent. It's admittedly one of my least favourite parts of the process—what do you mean someone else has to read and perceive the things I've written, what the hell—and the thought of even being present in the document while someone is scrutinising my work makes me want to vomit.

Which is why I don't know whether to applaud or be terrified of YouTuber Beluga, who invited 833,013 people to edit a document titled: "A Hermeneutic Exploration of KSI's 'Thick of It' and its Multilayered Significance'. Which, yeah sure, okay.

The video Beluga posted to YouTube shows him drafting a message that mentions every single person in his server—today I learned that Discord warns you before you try to ping nearly one million people—which read "my school project is due in 10 mins. pls help!!!!!"

Unsurprisingly, it takes absolutely no time at all before over 100 people have already started to flood into the document. It seems like a few people start by trying to help out with some genuine edits—or at the very least make very light changes—before one bozo comes along and simply. Deletes the entire document.

It doesn't stay blank for long though, as it begins to rapidly move through a chaotic series of pasted messages, memes, and nonsensical ramblings. I was able to find people spamming keys to craft Shakespearean scripture such as "HERNOggyjbbjutynubkrnuknukeunf". You know, the good stuff.

I do also, unsurprisingly, spot a slur—which Beluga has dutifully blurred out—and chantings of "HELLO BELUGA". It's after this where things start to get pretty… weird. I get flashbanged by a glimpse of a pregnant emoji, an ASCII art toucan, a horrifyingly realistic minion, before I get flashbanged again by someone writing the word BALLS in a huge font.

The doc ends up growing and growing to over 600 whole-ass pages, before Google Docs craps out altogether and throws up a nice error message for Beluga. Overall though, the one-minute video is, for the most part, a surprisingly pleasant mix of wholesome and cursed. At one point, one Google Doc-er calls it the "least chaotic Beluga video"... before being consumed by "SIGMA SKIBIDI DANCE" pasted over and over again.

Glancing over Beluga's other videos though, it defo seems like it might rank a lot lower on the chaos scale than I thought after watching it. The YouTuber's recently uploaded videos like "Giving 841,709 people my Netflix password," "Inviting 851,309 people to a Zoom meeting" and "Wishing 793,159 people a good night". An agent of chaos, both big and small. This is what happens when people remember they have free will.

Some people look at a 3D printer and think "this will make useful parts for my projects." Some people hear an air raid siren, and assume there must be an historical war re-enactment taking place nearby.

I am the third type of person, who hears a 3D-printed air raid siren and experiences a flash forward to my apocalyptic future yet to come. Bear with me here folks, I'm cooking.

Youtuber Mark Makies has created a 3D printer air raid siren. This you probably know, if you viewed the video above (via Hackaday). What I'm not entirely sure of is why it sounds like both a comedy bit in once-popular children's TV show Noddy's Toyland Adventures, and the sound of impending doom, all at once.

It's a cool project, anyway, and I'm all down for that. A combination of cheerfully-coloured 3D-printed components and RC parts have been employed to make something both charming and terrifying, and for this Mark should be applauded.

Internally, a brushless motor controlled by a speed controller turns a rotor, which pulses air pressure inside a stator. The slots inside the stator (stay with me, this is technical) make a wooooooo sound through the honking trumpet, and cause the hairs to stand up on the back of my neck. Something like that, anyway.

Impressive. A potentiometer attached to the speed controller means the pitch can be adjusted by ramping the speed of the motor up and down, meaning those wooooos can become weeeees, and so forth. And for some reason, it reminds me of my youth.

I grew up in a small town just outside London, and every Tuesday morning a WW2 air raid siren would go off, waking me from my slumber/hangover. To this very day, I know not why. Friends would discuss it over an afternoon pint. Why does the siren sing for thee?

Still, it was both a useful way to tell me I was late for work, and a good excuse to check the skies to make sure the end times had not begun while I slept, blissfully unaware of my potential fate. The world kept turning, and the siren, it seemed to sing for no reason at all.

This video takes me back to those times. Ah, the halcyon days. Now, I fear that one day, I will hear the siren again. Hopefully it's a brightly coloured one, at least.

I'm old enough to remember a pre-YouTube era, where children played happily in the fields and everyone wore fetching hats to church. Now, though, it's a staple of so many of our lives, mine included. And according to YouTube CEO Neal Mohan, it's not just dominating the arena of our phone and desktop PC video watching, but on track to take over television, too.

In a blog post on the, err, YouTube official blog, Mohan takes a moment to mark the internet video sensation's 20th birthday, with his four "big bets" for YouTube in 2025 (via Sweclockers). "YouTube will remain the epicenter of culture", he says. Heavens help us all.

"For more and more people, watching TV means watching YouTube. Viewers are watching, on average, over 1B hours of YouTube content on TVs daily, and TV is now the primary device for YouTube viewing in the U.S" says Mohan.

"It’s interactive and includes things like Shorts (yes, people watch them on TVs), podcasts, and live streams, right alongside the sports, sitcoms and talk shows people already love."

I feel like following a statement on Shorts viewership with a "yes, really" qualifier is perhaps a tad defeatist, but I'll admit that even I watch the occasional YouTube Short. On my TV, though? That's sacrilege, surely.

In other news, YouTube TV apparently has more than eight million subscribers and YouTube Premium has over 100 million happy payees. Given YouTube's increasingly aggressive (and incredibly lucrative) policy of filling my watching hours with ads, I've considered paying for one myself on occasion.

Ah, who am I kidding. That'd eat into my Steam budget, and I feel its kinda like letting that sort of ad-based incentive win. Anyway, Mohan also boldly states that "YouTubers are becoming the startups of Hollywood," as would-be filmmakers are starting off on the platform in the hope of graduating to the really, really big screen:

"Creators are bringing that startup mindset to Hollywood: leaning into new models of production, building studios to elevate their production quality, and exploring new creative avenues.

"We're committed to meeting creators where they are with tools and features that power their businesses and communities" he continues. "We’ll continue to support their growth through more traditional revenue streams like ads and YouTube Premium, while introducing new ways for creators to partner with brands to bring their products to life."

Ah good, more brands. It seems like most of my favourite creators are now acting like the QVC shopping channel and hawking dubiously-effective wares, although I've not yet been tempted to buy anything simply because a gurning thumbnail-enthusiast has shoved it in my face mid-video.

Still, it must be working in general. There's gold in them thar hills, or so I've been told.

And actually, while it's fun to poke fun, I think Mohan has a point. Even my dear sainted mother has been known to watch YouTube on her televisual box, and if that's not a sign of mass-adoption, I don't know what is. YouTube is on your phone, your desktop browser, and now, it seems, increasingly replacing the other apps on your TV set-top box.

A brave new world indeed. Now, if you'll excuse me, I'm off to tie a ring of daisies in my hair and frolic among the heather. It's a beautiful day outside, y'know?

There's not much we can agree on in this world of strife and tears, but there is one opinion that's just about universal: YouTube ads suck. Want to watch a video? Watch this ad first. Want to watch another video? Watch this longer ad! Can you click past the ad before it's over? Maybe—pay close attention to find out. As for ad blockers, YouTube is doing its best to make those more hassle than they're worth. It's obnoxious as hell, and it sucks.

Given how offputting and utterly detrimental to the experience its ads are, you might fairly wonder why YouTube keeps shovelling them on. Why does it work so hard to make YouTube worse? The answer is very simple: Those hated YouTube ads make, as they say in the Canadian scientific community, a metric ass-ton of cash.

Alphabet's Q4 financial results (via Variety) reveals that YouTube ads raked in more than $10.4 billion in the quarter—to be clear, that's 10 billion smackers in just three months. That's a relatively small slice of the total Alphabet pie, which topped $96 billion (again, in three months), but it's a staggering amount of money for three months of ads on the 'Tubes. It's also a new record, marking a 14% increase over YouTube's advertising revenue in Q4 2023, which hit $9.2 billion.

In its earnings call, Alphabet CEO Sundar Pichai attributed much of that jump to the US election, which took place on November 5, 2024. Combined spending on YouTube ads by the Democratic and Republican parties was almost double what they spent in the 2020 elections, and more than 45 million people watched "election-related content on YouTube" on election day alone.

It's possible to dodge the deluge of ads, of course, by signing up for the YouTube Premium subscription service. That goes for $14 per month or $140 per year, which strikes me as an exorbitant amount of money for, y'know, YouTube. It'll be a cold day in hell before I throw that kind of money at, again, YouTube, but at least a few of my compatriots here at PC Gamer—like associate editor Ted Litchfield—do pony up for it, and I'm ashamed to say they all claim it's worth it.

They're not alone in that assessment, because it's raking in the green too. YouTube subscription revenues aren't reported separately but are instead rolled into "Google subscriptions, platforms, and devices," but that's up year-over-year as well, from $10.8 billion in Q4 2023 to $11.6 billion over the same period in 2024. Alphabet chief financial officer Anat Ashkenazi said during the earnings call that "we continue to have significant growth in our subscription products, primarily due to increase in the number of paid subscribers across YouTube TV, YouTube Music Premium and Google One." YouTube TV, for the record, is a separate streaming service, unrelated to YouTube Premium, that starts at $83 per month.

Oh, but don't worry, because it'll probably get worse. "We believe that AI will revolutionize every part of the marketing value chain and, over the past quarter, we’ve seen how our customers are increasingly focusing on optimizing their use of AI," chief business officer Philipp Schindler said. "As an example, Petco used Demand Gen campaigns across targeting, creative generation, and bidding to find new pet parent audiences across YouTube. They achieved a 275% higher return on ad spend and a 74% higher click through rate than their social benchmarks."

Schindler said later that, based on Nielsen analysis, "Google AI-powered video campaigns on YouTube deliver 17% higher return on advertising spend than manual campaigns."

So there you have it: Everyone hates YouTube ads but they make a mint, so we're stuck with them and can probably expect them to burrow even deeper into our psyches in the future unless you're willing to spend 14 bucks every month to make them go away. Maybe that's not such a terrible price to pay after all.

I have precisely one tattoo, which I paid a trained artist an exorbitant amount of money to ink into my pasty skin. YouTuber Emily The Engineer may have developed a more cost-effective solution, however, as she's cracked open a 3D printer, bypassed its firmware, equipped it with a tattoo gun—and even found a willing volunteer to test it out.

Don't worry about the test subject's safety, though—there's a handy machine-halting panic lever installed for when they inevitably realise they've made a terrible mistake.

As Emily points out, she's printed a "ton of stuff" over the years, not least this world's largest Benchy boat attempt (via Hackaday). However, "printing" on a human is a brave new frontier, and began with the merciless tearing apart of an old 3D printer frame capable of mapping out the X and Y axis of a 2D design.

Then began a process of convincing the printer to ignore things like temperature and medium variables, before a few 2D pen sketches were drawn on test paper and the back of Emily's hand.

A small amount of Z-hop was added in between strokes to prevent unintended lines, before the printer was deemed ready for the inclusion of a tattoo machine.

To add to the fear factor, Emily admits that the first time she ever touched a tattoo machine was for the purposes of this build. Still, it wasn't long before the machine was installed on the print head, and a rough design was inked onto practice material.

A problem soon became apparent, however. The machine can only "print" accurately on flat surfaces, whereas Emily's (somewhat) willing test subject, Dan, wanted a design on his leg. Legs are not flat. So rather than re-engineer the entire apparatus, a bracket was made to flatten out the top of Dan's thigh to squish it into a flat surface.

A lever-controlled solenoid was also added, which lifts the entire tattoo machine vertically off the surface if something were to go horribly, terribly wrong. Safety, folks. It's important.

One last (slightly wonky-looking) material test later, Dan bravely put his leg in the bracket and strapped the tattoo-3D printer hybrid onto his thigh—before the needle finally pierced the top layers of his skin. According to Dan, the sensation "wasn't bad", and he's now the proud owner of... well, this:

Good job, everyone. I've certainly seen worse tattoos (some of them on close friends, you know who you are), although it's not quite the "Rico-at-the-end-of-Starship-Troopers" futuristic tattoo machine we all had in our heads.

Still, I salute the attempt. I've seen 3D printers co-opted into all sorts of weird experiments before (this chocolate-printing machine is a highlight), but a full-on automatic tattoo machine is a new one.

And as for Dan? Well, he'll now be able to stare down at his right thigh for the rest of his life and know exactly what he's looking at. Rather you than me, Dan. Rather you than me.

Seven months after demonetizing Dr Disrespect's channel over his admission of inappropriate communications with a minor, YouTube has decided that Guy Beahm, better known as the streamer Dr Disrespect, has served his penance and can start monetizing his videos again.

After multiple sources reported the change in direction, Beahm himself triumphantly announced the remonetization in a message posted to X:

YouTube confirmed its decision in a statement provided to PC Gamer. "Dr Disrespect was previously suspended from the YouTube Partner Program for violations of our Creator Responsibility policies," a YouTube spokesperson said. "Creators who are suspended from this program can reapply for access, and after careful review of the channel’s recent activity, we’ve reinstated it. If there are further violations, we’ll take appropriate action."

Content creators whose channels are demonetized are allowed to apply for reinstatement after they've addressed the content or behavior that resulted in the suspension, according to YouTube, and a green light was given in this case after the aforementioned "careful review" of Beahm's channel activity. But Beahm wasn't demonetized for his channel content, but for violations of YouTube's Creator Responsibility policy, specifically the allegations—which, again, he confirmed—of inappropriate DMs with a minor.

The natural question, then, is: How has he addressed that? Aside from acknowledging that the allegations were accurate, Beahm's only public actions have been to deny any illegal activity and paint himself as the victim of a behind-the-scenes conspiracy at Twitch, where he remains banned. There have been no further claims of inappropriate behavior on Beahm's part that I'm aware of, so I suppose one could call the situation "addressed," as far as it goes.

But carrying on with business as usual while pinning the blame for your misfortunes on others and insisting you didn't really do anything illegal does nothing to "address" the initial violation, and that leads me to question YouTube's real commitment to its policies. Saying you want to "protect the YouTube community" with your rules is nice, but remonetizing a streamer who makes a point of showing no contrition for his behavior sends a very different message. Twitch felt Beahm's actions were egregious enough to show him the door permanently; YouTube's response pales by comparison.

Of course, one might also note that Beahm remains a major draw, and is attracting a significant audience on YouTube competitor Rumble, which he's been actively promoting on his social media channels. Despite everything, where Beahm goes, an awful lot of viewers—and money—follow.

In a bizarre twist of timing, Beahm's return to monetized YouTube happened on the same day that Midnight Society, the studio he co-founded in 2021 with former Call of Duty creative strategist Robert Bowling, and Halo 5 multiplayer designer Quinn Delhoyo, announced that it is closing down without releasing its in-development extraction shooter Deadrop. Midnight Society parted ways with Beahm in June 2024 after the allegations of his inappropriate messages first came to light.

Faceless AI-generated YouTube channels have become a problem as of late, so one tech YouTuber took it upon themselves to poison the data they are trained on.

In her latest video (via Ars Technica), f4mi has shown off a way to mess with the data that AI is trained on, specifically the subtitles of YouTube videos. Effectively, these bots can scrape all the information in the video by accessing the data in subtitle files, which are then synthesised with all the other YouTube videos it has taken data from.

Someone can then instruct that bot to create a new video in the style of other videos. The tools aren't quite advanced enough to animate a lifelike face and the creators won't put their own face on it. That is where the "faceless YouTube channel plague" (as popular YouTuber Kurtis Conner describes it) comes from.

"All this AI slop that's appearing on every social media platform is not made by robots trying to steal our jobs," says f4mi, "it's made by humans trying to make money using AI to launder other people's work."

A little while back a life hack popped up online. Job applicants were putting positive keywords in white between margins in document folders to trick AI bots into thinking they were more employable. The poisoning method that f4mi invented is sort of like this.

It grabs the standard subtitle text for a video, adds gibberish in between the lines, makes it invisible to the average user, and then feeds it back to YouTube. She did this by putting works in the public domain into her files and using a tool to make synonyms for many of the words so the AI doesn't spot what she's doing. A Python script simply went through and made these changes.

Unfortunately, tools that can transcribe audio get around this but a few large LLMs like ChatGPT can't. Instead, the AI would 'hallucinate' inventing ideas about the video based on what it can grasp from the enormous amount of text in its subtitles.

And this is where the '.ass' subtitle format comes from. It can be used for standard subtitles but also for animation and graphics, which allows the level of customisation needed to sneak in all of that pesky extra data. Entire books can be hidden with the file type's position tool to make both the size and transparency of unneeded text zero.

.ass is not among the supported files for YouTube videos—but you can convert an .ass file to an SRV3 file, which will then keep all of the data from your original file.

Unfortunately, f4mi's video had problems with phones crashing "due to the subtitles being too heavy". YouTube has regional subtitle files you can activate, which allows users to account for cultural differences in phrases and tone, so the poisoned track only exists in the UK subtitle track for now, which is the original .ass file. f4mi has uploaded standard files for Australia, which means AI summary tools are getting around the poisoned track and to an untouched one.

This means that plugging the video into an AI summary tool now works. However, as a proof of concept, this is a very smart workaround that means users wanting to steal your videos may have to do it the old-fashioned way, which might just be enough to deter many.

f4mi is one of a long line of YouTubers (like Drew Gooden and Hank and John Green) who have been fighting against AI plagiarism on the platform, and likely won't be the last. It's unlikely the next major anti-AI effort will have quite as good a file name attached, though.

Just months after laying off a "significant" number of employees in the face of "unexpected challenges," Midnight Society, the studio co-founded in 2021 by Guy "Dr Disrespect" Beahm, former Call of Duty creative strategist Robert Bowling, and Halo 5 multiplayer designer Quinn Delhoyo, has announced that it's closing completely.

"Today we are announcing Midnight Society will be closing its doors after three incredible years, with an amazing team of over 55 developers contributing to our new IP Deadrop," the studio said in a message posted to X. "We are actively seeking other game studios that would be interested in offering employment opportunities to our talented team members. If you know anyone who's hiring please forward this message to them or DM us for direct intros.

"We express our sincere gratitude to each and every one of our community members and [are] deeply sorry we were unable to reach our ultimate goal."

That goal was the release of Deadrop, an extraction shooter that got its start through the sale of NFTs. The money sunk into those now-homeless digital doodads has led numerous "founders" to ask in response to the message if they'll receive refunds, since the game isn't happening. Midnight Society hasn't officially commented on the matter, but the safe bet (and general consensus) is, "no."

Midnight Society came out with a bang, backed by Beahm's fame as a streamer and Bowling's cred as a Call of Duty veteran, but progress on Deadrop seemed to slow in early 2023 when the studio moved away from frequent "snapshot" build releases to "less frequent but much larger" updates. Things got ugly a year later when Beahm was fired by Midnight Society over allegations that his lifetime ban from Twitch was issued over an exchange of inappropriate DMs with a minor.

PC Gamer's Morgan Park said at the time that "it's hard to imagine a happy ending for Midnight Society," words that now seem prescient.

In a twist of horrible irony, Midnight Society's closure comes on the same day that Beahm's YouTube channel was remonetized. It was demonetized after his admission of the inappropriate message exchange with a minor; his first request to remonetize the channel, in October 2024, was refused; YouTube's appeal process allowed him to reapply for monetization again after 60 days.

I've reached out to Midnight Society to ask about the possibility of refunds for founders, and will update if I receive a reply.

Half-Life 2, Halo 3, Fallout: New Vegas, and Knight Rider: The Game. All iconic titles in their own right and all using Havok's iconic physics or pathfinding engine components. If you haven't seen the legendary Havok saw logo in some time, a recent physics engine test video could signal a comeback.

Other than a presentation at the Game Developers Conference and another at Develop: Brighton last year, this is the first promotional video posted to the Havok YouTube channel in 10 years. The last was titled "Havok Powered Titles at GDC 2014" and shows off gameplay from Assassin's Creed: Black Flag, Battlefield 4, and Halo 4 among others.

The trailer itself is rather impressive. Starting with a blaring siren and presumed crisis, a figure stands alone in a dig site of some kind, before perfectly destructible rocks start to really test that physics engine. Stone bridges collapse, debris falls, and an electronic figure runs through the remains. Standard 'show off the engine' stuff.

However, Havok isn't finished showing off its fancy engine work here. The trailer continues, showing skulls clattering on Indiana Jones-style spike traps (sorry I can't stop thinking about the new game), and fire erupting through the ancient halls.

Havok offers three central products to videogame studios: Havok Physics, which deals with real-time particles, collision detection, and more; Havok Cloth, which adds physic-based movement to character clothing; and Havok Navigation, which hands NPC intelligence and pathfinding. Havok Physics is what is being showcased here and the trailer does a good job of doing so.

Despite me name-dropping older games above, Havok has worked with plenty of newer titles, like Helldivers 2, Sonic X Shadow Generations, and The Legend of Zelda: Tears of the Kingdom.

A video like this, due to the fact that Havok doesn't publish or create its own games, is more likely to be used in pitching to developers, rather than gathering any interest from the general public. This could explain why the YouTube channel has remained largely inactive for some time. The new video coincides with the launch of Havok 2024.2, which upgrades all three of Havok's products.

However, it's still nice to see what such an iconic software studio is up to and the dynamic destruction and lighting look great. Truly destructible worlds are a rarity, and as the tech surrounding games only gets more advanced, we could hopefully see the full implementation of this tech in future games.

Excuse me while I go locate my old copy of The Lord of the Rings: The Third Age. The nostalgia has got the better of me.

If you fancy yourself as being the next tech YouTuber waiting in the wings to explode onto the scene, you might want to pause for a moment and reflect on the amount of time you'll need to spend generating all those rage-baiting thumbnail images of your face to add to your content. But fear not, as there's now an AI tool that'll handle all your gurning goals for you.

The tool in question is called Expresso because…umm, I'm not sure why, to be honest, but it's by EasyStudio AI which also offers an AI for editing and tweaking photos. I should imagine that there was a meeting of managers, at some point, that wondered what else its servers could be used for and they all settled on creating something to make custom headshots.

Although the tool's interface isn't overly intuitive, it is simple to use—just upload a headshot photo or image, and then select a preset for a particular expression you want the face to have. You can adjust the positions of the head, eyes, eyebrows and mouth to get that perfect 'smile that never reaches the eyes' or 'ermahgerd it's so bad' look that's pretty much the norm on every YouTube thumbnail these days.

Being the inquisitive and top investigative journalist that I am, I signed up for a free account (which gives you three goes at generating an image, in the form of credits) and set about seeing just what AI could do for me. Before we get on to how insufficient three credits are at road testing something that would otherwise cost $10 a month for 100 credits, let's see what I managed to create.

Look on my works, ye mighty, and despair! It's almost as if Percy Shelley was a 19th-century AI prophet, rather than some genteel British poet. Anyway, I have to say that Expresso somewhat lives up to its name, simply because it spits out an image very quickly, though I dare say that if it becomes super popular and the servers clog up with requests, it might not be quite so speedy.

However, my issue with it is that there's no kind of preview system and you will use up one credit to generate any image. If you're a full-time content creator and you're looking to create the perfect thumbnail clip, then you could well burn through a whole stack of credits just for one headshot.

The first AI-generated image I created with Expresso (second in the gallery above) was with the standard 'surprised' preset but the final two are tweaked versions of 'shocked' and 'confused' respectively. I don't think I'm alone in thinking that none of these facial expressions (oh wait, that's why it's called Expresso…express-o-ions…) are good approximations of those emotions, but at least they look realistic enough.

But that's AI to a tee and anyone who's spent time with an AI image generator, such as Stable Diffusion, will know that you have to spend a lot of time trying small changes over and over again, to get the exact image you want. In the case of Expresso, that means a whole lot of money.

I'm clearly not the target audience for this AI tool but I'm not overly convinced that any professional YouTuber will spend good money on this when it's not really all that hard to have someone take a few shots of you with a camera and then quickly edit them in a free photo editing app.

AI doth giveth, AI doth taketh awayeth.

Zara Dar is an engineer who dropped out of a PhD to become an OnlyFans model and content creator. Her channels on both YouTube and Pornhub often share the same videos: math concept explainers that teach you what a neural network is or where the concept of pi came from, and which just happen to be delivered by an attractive woman in a low-cut top. As Dar explained on LinkedIn (via 404 Media), the same video will earn her $340 per million views on YouTube, and $1,000 per million views on Pornhub.

Admittedly, her videos get more views on YouTube because I imagine there are less people going to Pornhub to find out what integrals are. But there is a surprising amount of non-pornographic stuff on Pornhub, including loads of Minecraft let's play videos. Comedian Ryan Creamer has a whole channel of wholesome videos that just happen to have suggestive titles. (That link is still NSFW, by the way, unless your work is cool with you checking out Pornhub while you're on the clock like mine is. I hope.)

Dar told 404 Media that she's not entirely sure why her videos are doing so well on Pornhub, "but it could be because my SFW videos stand out against the typical NSFW content on the platform. That contrast might make them more intriguing or refreshing to viewers. But that's just my speculation."

After posting an explanation of her business model to LinkedIn, which briefly went viral, Dar's account on the website where business bros go to explain how great it is to ignore your family in favor of being a workaholic was banned. "I was engaging professionally," she said, "but LinkedIn's strict handling of this situation feels counterproductive to its purpose."

The first four episodes of Beast Games have now aired, the beginnings of what is likely to be a long-term collaboration between the world's most popular YouTuber and Amazon Prime. Beast Games is inspired by both Netflix's Squid Game and the most popular video on MrBeast's channel, "$456,000 Squid Game in Real Life!", and the basic setup is 1,000 contestants competing for a $5 million prize while being brutally eliminated via simple team games and in some cases psychological tests.

Squid Game is one of Netflix's breakout hits, with the second series out now, a fluorescent and creepy tale of what desperate people will do for money, and the people who exploit them and watch. The show's twist, which is a minor spoiler though it becomes clear early in the series, is that these childrens' games are deadly, and the losers do not walk away. One of the threads in the series is how different contestants deal with this realisation, and each other, while coping with the degrading and unsafe conditions they're kept in en masse between events.

Beast Games is aiming to be Squid Game without the literal death, but in the amount of money involved and how things start to play out the similarities and differences between the two shows comes into stark relief. Squid Game is all about what desperate people will do for money, the extremes of behaviour they can be tempted into, and the people who exploit them and watch and play on the greed. So is Beast Games.

Unleashed?

From the off Beast Games foregrounds two things: MrBeast himself, who in this slightly more traditional 'game show host' role is even more mannequin-like than usual; and the "life-changing record-shattering" $5 million dollar reward. The first episode then makes one of the big reveals that runs through Beast Games' whole setup: MrBeast and his flunkies (more on them later) have a lot more money than $5 million to give away, and nakedly bribing contestants is a quick way to thin out the crowd and get some easy drama.

As for those contestants, there is a certain kind of performative angst that underlies almost everything about how people present themselves on a show like this, the same behaviour you see from a child told to be extra-appreciative of their Christmas presents. Every announcement, no matter how banal, is greeted with someone clapping their hands to their head and exclaiming "oh my god" or "this is crazy!"

One contestant is there to win to help out the homeless. Another wants to give it to her community. Others freely admit they are in it to buy "gold chains" or just straight-up change their own life, and fair enough. These little moral moments are the stock-in-trade for most US reality/game shows, but are brought into stark relief by the sheer amount of money being thrown around and the show itself constantly pounding the drum about "the biggest cash prize ever!"

The first challenge involves, according to MrBeast, a "one meelion" prize that will be shared by any competitors that quit right now. It's not surprising that many choose to do so, 52 eventually choosing to go home with just under $20K each. It's grim to see a camera trained on a young woman's face, as she talks about what her family would think of her for turning down this sizeable chunk of cash—she'll go on to be eliminated and receive nothing—before it cuts away to the next victim.

These people are victims. There's no crime and of course they all signed ridiculous NDAs to get a shot at the loot, but this is just a glitzed-up equivalent of making paupers fight for scraps. Desperation is the scent of the whole show, as distinct as brie behind the radiator, and the way the contestants are manipulated amps this up: What is this worth to you? Imagine what you could do with this money! But then that is the secret behind MrBeast's seismic popularity: He got there through virality and stunts, but stays there by giving the people what we want. It is still bread and circuses.

The second game of the first episode is where things threaten to become nigh-unbearable for the sane viewer. The contestants are split into rows and told one member of each row must self-eliminate for the rest of the row to progress: The last three rows to have a sacrifice will be eliminated entirely from the game.

Could you put it more on-the-nose? The shrieking and bartering and second-guessing that instantly, brazenly follows from competitors (well, the ones they included in the edit) is just grotesque, a masterclass for students of insincerity and dotted with these sad moments where some soul is peer-pressured, by their grinning teammates, into sacrificing themselves "for the greater good" of people they don't know in a team that was formed minutes ago.

These people will do anything, these people will say anything: And some will say good for them. At the end the cameras capture one man screaming into the camera, neck tendons straining, "Yeah! We made it! We made it!"

If you think this can't get more nauseating, the losing contestants are then literally dropped through a hole: The panel they're standing on disappears, and they're yoinked down into the abyss while MrBeast says something like "you didn't think we could drop 80 people?" I have never thought that in my life and yet here MrBeast is, on my television screen, doing it.

It's during these tasks that MrBeast's gormless co-presenters are given most screen time and, honestly, all they do is yell things at people. The real ick factor though is their clear sense of self-superiority over the meat, something that literally comes from the top: As MrBeast looks down on the massed contestants in episode two he says “they look like ants.” There's a point later on where the crowd has to move towards MrBeast on the ground for the next task, and as soon as it does one of the flunkies blurts out "it's scary". MrBeast considers this, and responds "it's like a zombie horde".

The set is both expensive-looking and grotesque, with the centrepiece for the second episode looking like a multi-storey car park bisected by a giant elevator shaft. MrBeast and flunkies ride up and down between the four floors, and at points even take their scissor lift close to the peons for some words of condescension.

What’s odd is how cheap the games themselves feel next to all this set dressing and the obscene amount of money being thrown around. The first game in episode two sees both teams catching balls that drop from multiple holes in the roof, which no matter how many effects you overlay in post-processing is hard to get too excited about. The most fun thing about it is the ludicrous attempt to make this gaudy spectacle seem sinister by having two characters dressed like dime store versions of Squid Game’s Front Man dropping the (bright red!) balls into a pipe.

These characters hang about and are used for various establishing shots but it’s all a bit awkward because there’s nothing for them to do, really, beyond some part-time event work. MrBeast and the flunkies are in control and set the tone, histrionic and narcissistic to the core.

The chintzy feel of the games persists. The second game is yet another self-sacrifice challenge where contestants have to give themselves up so their team survives. The “twist” is a pair of curtains separating the teams (so they can’t see how many of the other team have sacrificed) alongside a phone line connecting them. Two episodes in and this is already repeating a game concept and adding little twists to justify the formula. Things do pick up in later episodes, but this is not quite what you expect from the most expensive series ever made.

MrBeast’s major sin as a presenter is that he doesn’t have a discernible sense of humour. There are lukewarm gags scattered throughout his script, but the guy’s only delivery is a bark and he can’t even manage decent banter with the guests. One of his favourite words is “insane” and, yes, a lot of what you’re watching probably does fit that definition: But he doesn’t mean it that way, and lacks the wit and self-awareness to see why others might.

That’s the most baffling thing to me about the MrBeast phenomenon. There’s a wildly successful persona there, but just no personality: The man is not for us to know beyond a gurning cipher.

MrBeast stands on that pile of money while contestants are "screaming at him like he's a god", but the real God is the money. It is what everyone genuflects towards, it is $5 million, the alpha and the omega, with contestants dropping like flies as bribes are offered. But there are surprising moments: at one point four team leaders refuse a bribe that rises to a million dollars, which would give them the cash and a pass to the next round but doom their dozens of teammates. Watched on by the people they’d screw-over, and goaded by MrBeast and crew, none of them take it and every team goes through.

Reality bites

In the next round one of those “heroes” is eliminated after he gets an unlucky sandwich. He turned down a million dollars and goes out with nothing. But this is what we're here for really: Not the winners but the losers, the awesome devastation that MrBeast and his crew can unleash with numbers, the capriciousness and cruelty of fate being visited on others.

"This is diabolical!" shouts one contestant in glee. Well, yes, but maybe not in that way. I always thought the name was stupid but maybe MrBeast as a persona makes some kind of sense, the human touch of the pronoun framing animalism.

There's a sequence where the contestants are set up to take part in a sack race. As the games go this one is relatively low-stakes, but then shows the contestants bumping around and falling over to classical music in slow motion, aping a signature Squid Game technique: As if ripping off the concept wasn’t enough.

It only accentuates the differences. The parts in Squid Game where the contestants see their accumulating prize always come after a scene of horror, with the various reactions lingered-over and, at times, the mere thought of the money inspiring terrible subsequent acts. In Beast Games, you just get repeated shots of MrBeast and piles of money and people screaming about it. Then the timer starts, the cameras roll, they grasp for it, and we gawp.

There are those who will rightly point out that Beast Games isn't really doing anything new for a game show. At bottom these things are about people competing for prizes of one sort or another, and the only new thing here is the scale that Amazon enables. But that scale is so breathtakingly large thanks to Amazon's millions that the culture it's celebrating is stark, the YouTube fantasyland of "epic" giveaways and fratboys showering dollars on the great unwashed, each contestant a stick to beat the others with.

Beast Games is a unique product of our times, a mirror to a society rife with excess and disparity, where the haves say "jump" and the have-nots says "how high?" But I think what makes it stick in my craw is the presentation of all of this as aspirational, like it's some sort of life achievement to become the lab rat pulling levers in hope of a reward. That's entertainment, I suppose. So why am I not entertained?

Boxing, and boxing games, are in a strange place at the moment. In the sport itself, we have a high-ranking member of the Saudi Arabia Royal Court clicking his fingers like a Roman Emperor at the Colosseum to create the fights that fans have long been wanting to see. On the games front, meanwhile, we have Undisputed, a flawed but strongly supported boxing sim that's undisputed in its genre by nature of the fact that it's the only serious boxing sim to have been released in over a decade (and, if I'm not mistaken, the first ever one on PC).

Another thing that speaks to the state of boxing is that the headline feature of the new Undisputed DLC released yesterday isn't the three Mexican boxing legends included in it, nor James 'Buster' Douglas—the man who infamously toppled a prime Mike Tyson in one of the biggest upsets in the history of the sport. No, front-and-centre of the new DLC is the man who beat a 60-year-old iteration of Mike Tyson, 'The Problem Child' Jake Paul.

Now, I don't doubt that there are plenty of people who'd happily fork out the $16/£13 asking price just to send the YouTuber reeling around the ring with haymakers, but it might not be so easy. Based on the 12 fights he's had against retired MMA fighters half his size, basketball players, and a boxing legend some 35 years removed from his best days, the devs have deigned to give Paul a rating of '84'. That makes him better than Hall-of-Famer Arturo Gatti, on par with two-weight world champion Ricky Hatton, and good enough to give the 87-rated Light Heavyweight version of Mexican superstar Canelo Alvarez a decent scrap.

On the bright side, a big free update has come to Undisputed alongside the paid DLC, and includes a couple of new venues, improvements to the career mode (you can now move up and down in weight classes), and various other tweaks. What the update doesn't seem to address however, are ongoing issues with the game's unstable online mode, nor generally unsatisfying in-ring mechanics. Having last played the game about a month ago myself, I found that punches felt far too soft, were massively inconsistent in registering, and there was an overall lack of fluidity when it came to stringing together combos. It's kind of stodgy, and feels inferior to Fight Night Champion, which came out 11 years ago now and is in my eyes still the GOAT of boxing games.

It's a shame, really, because Undisputed's popularity speaks to how much demand there is for a Fight Night successor, but at the moment it feels like licensing and collaborations with Saudi Arabia and Jake Paul are a higher priority for the developers than getting the game itself fighting fit.

Have you heard of the dead internet theory? It's essentially the idea that the internet is primarily populated with bots and AI-generated responses, intentionally designed with the overall goal of minimising human interaction and promoting products. The theory posits that humans on the internet are relatively rare, but bots responding in a manner just like them are everywhere. Including YouTube.

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.

Sometimes, certain questions keep you up at night. For me, it's how much longer I have on this planet before something dreadful happens to me, because I'm a happy-go-lucky soul. For YouTuber Nate Scovill, however, the question burning a hole in his brain was much more straightforward: Which Lego hurts the most to step on?

Deciding there was nothing else for it but to get scientific, Scovill set to work building a test rig to find out (via Hackaday). Of course, they could have chosen to step on some Lego themselves to judge the results, but as they put it: "I didn't want to show my feet on camera. If I'm going to end up on Wikifeet, it's gonna be on my own terms."

In fact, there's a whole Wikifeet sideplot going on in this video, but I'll let you watch it for yourselves to find out where that goes. We're concerned with science here people, and so Scovill started out with a system capable of approximating his own weight, with an arm that would slam down upon the offending Lego brick with cardboard squares attached to the end to record a "gradient of damage."

Initial results were promising. A little wizard's hat delivered surprising amounts of penetration, thanks to its pointy top and stable base. However, nothing quite scared Scoville like an old Lego wheel, now out of production supposedly because of its potential danger to the underside of your feet.

On one side of the wheel is a metal axle, that when placed face up punched a hole into the cardboard test rig. Scoville also points out that the metal is prone to rusting, adding "toxic damage".

However, Scoville didn't stop there. A ballistics gel foot was created, inspired by Mythbuster's judicious use of the material. Unfortunately, the zip ties on the end of the test rig ended up damaging the foot in unintended ways, meaning that Scoville was left with little choice but to put the foot on top of the Lego piece and slap down the bar.

Scientific method, destroyed. Well, not really. The ballistics gel foot proved massively vulnerable to the terrors of pointy Lego, leaving some embedded underneath in a way familiar to any of us that have jumped around the front room on a Sunday morning with a small piece of a child's toy stuck in the bottom of our soles.

I'll let you watch the video to see the eventual conclusion. Still, those of you not from the UK may be unaware of something even worse you can step on in a dark room: A standard issue upturned UK plug.

Just looking at that image makes me wince. Anyway, I salute you, Lego science extraordinaire. You've gone where others fear to tread.

What's the most shocking bill you've ever got in the post? Personally, I'm still financially recovering from winter 2022, when a dalliance with an electric heater landed a fat £300 invoice on my doorstep. I write about videogames and subsist on coal soup, and there was Octopus Energy sending me a bill it presumably intended for Jeff Bezos.

Could be worse, though. I could be the guy at Google who had to open its $2.5 decillion fine from the Russian government (via The Moscow Times). Yes, that's decillion, which is a one followed by 33 zeroes. That means Russia wants Google to pay it $2,500,000,000,000,000,000,000,000,000,000,000. The number in rubles is even more absurd: ₽2 undecillion, or two followed by 36 zeroes.

For reference, the World Bank estimates the sum total of world GDP last year at $105,000,000,000,000—or one hundred and five trillion dollars. In 2017 (a while back, but not so far back that the numbers will be radically different today), MarketWatch generously valued the sum total of anything you could conceivably call 'money on Earth'—including cryptocurrencies, above-ground gold supply, and funds invested in financial products—at several quadrillions.

If you take out all the stuff that isn't actually money (so all the crypto, gold, and financial nonsense) that becomes a paltry $90.4 trillion. Regardless of which estimate you go with, $2.5 decillion is many orders of magnitude bigger. Almost incomprehensibly big. In other words, Russia wants more money from Google than actually exists on Earth.

Meanwhile, Google made $307 billion in revenue last year. If it tried to pay the fine using all that money—something it very much could not do even if it wanted to—it'd be a bit like trying to pay off your mortgage with a dime and two cents.

You may wonder how Google came to owe this ungodly sum to the Russian state. Per RBC it's because, erm, YouTube blocked some Russian channels. In 2020, pro-Kremlin media channels Tsargrad and RIA FAN (part of the Patriot Media Group formerly headed by Yevgeny Prigozhin, a stalwart Putin ally before he attempted to launch a coup last year and subsequently exploded in mid-air) won lawsuits against Google after the latter blocked their YouTube channels. The punishment? Daily penalties of ₽100,000, which doubled each week.

Like grains of wheat on a chessboard, the fines soon hit such phenomenal levels of exponential growth that it became impossible for any entity on Earth—or even every entity on Earth all together—to pay the total off. Fortunately for Google, I think it never had any intention of doing so anyway. The company ceased advertising in Russia after it invaded Ukraine in 2022, while its Russian subsidiary was declared officially bankrupt last year. If Google ever wants to do business in Russia again, I suspect the corporation and the Russian government will have to find some kind of accord to sweep the fine under the rug. Either that, or Sundar Pichai needs to start checking under the couch cushions.

Note: The sheer size of this number seems to have broken our website headlines, meaning I had to go in and manually disable our own sidebar to make it display properly. I consider this a feature rather than a bug.

Being candid with you, I checked out of watching Netflix's The Witcher TV show just a few episodes into season three. I'd felt the first season was, despite being notably flawed, an alright first stab at the dark fantasy Wiedźmin world of Andrzej Sapkowski, but after being served up similar fayre in season two and then, in my opinion, even weaker sauce in season three's opening, I lost interest.

And while I was disappointed, sure, I wasn't mad, because as I wrote about at the time a no-budget YouTube The Witcher fan film, watchable in its entirety for free, captures the vibes of The Witcher better than the Netflix show in my opinion anyway. The movie, which is named Alzur's Legacy, revolves around Geralt's fellow witcher Lambert, of Kaer Morhen's School of the Wolf, as well as sorceress Triss Merigold, as they hunt down a renegade sorceress from Aretuza who has stolen the infamous mage Alzur's Almanach.

Yes, Alzur's Legacy is a little rough around the edges and, yes, it is delivered in Polish with subtitles (although that's something that I feel adds to its flavour), but on a fraction of the budget of Netflix's show it somehow still manages to capture the flavour of the books and, yes, even the Witcher games, far better for me. I'd say it's definitely worth a watch if you're a fan, and especially so if you're disappointed to hear the news that it is now not just Henry Cavill who has dropped out of The Witcher Season 4, but also Kim Bodnia, the actor who played Geralt's Witcher mentor, Vesemir.

The news, which was first reported by Redanian Intelligence and then confirmed by Radio Times, means we will now have two new actors in both these lead Witcher roles in season 4. Liam Hemsworth has been confirmed as the actor replacing Henry Cavill as Geralt, but right now we have no word on who is going to replace Kim Bodnia's Vesemir. According to reports, Bodnia dropped out due to scheduling conflicts, while Cavill has reportedly moved on to further projects, including starring in a rumored Warhammer 40K cinematic universe.

In more exciting news for Witcher fans, though, the first new Witcher book in over 10 years is launching shortly in Sapkowski's native Poland, before being slated for an English-language release (as well as other languages) in 2025. The story and setting for this new book remain a mystery right now, however, in an 2018 interview Sapkowski said that it would likely be a "prequel or sidequel" rather than a follow-on to The Lady of the Lake, the final Witcher story in the official canon series' timeline.

If you fancy getting in the mood for this new Witcher book by watching Alzur's Legacy then I've got a tip for you. While you can of course just watch the movie on YouTube, the entire film is actually available to download in much, much better quality. And, what's more, you can download a wide series of subtitle packs for it, too, meaning that no matter the language you want to read, you've got the option available.

Lastly, if you are waiting for the next season of The Witcher on Netflix (no shade thrown, seriously, I'm glad it is working for you), then it looks like you still have a little time to wait yet, as the show is reportedly still filming, with four out of eight episodes now in the can. A 2025 release date is a near-certainty, though, and if the show continues to follow the release schedule it did for season three, then a summer 2025 release date seems likely. Who knows, maybe this changing of the guard for two of the show's lead characters will spark wider changes, too, injecting the show with, in my opinion at least, some fresh impetus.

Like many in Asheville, North Carolina, retro PC enthusiast Clint Basinger⁠—better known by his handle on YouTube, LGR⁠—was not expecting the catastrophic extent of Hurricane Helene's damage to the town and western North Carolina at large: "I thought I was decently prepared for what was to come!" Basinger told me via email. "We've had several tropical storm remnants pass through before, so it felt rather routine."

Through his channels, LGR and LGR Blerbs, Basinger has chronicled his extensive collection of retro PC games, hardware, and other paraphernalia to an audience of nearly two million subscribers. I previously wrote about his beige Pentium 3 tower stuffed inside a giant Ikea Teddy bear, while some of my other favorites include a PC built inside an evil glass pyramid, a 2003 PC with a faux fish tank viewing panel, and his long running series on goofy computer mice. The thing I've always appreciated is Basinger's infectious enthusiasm for the material: Rather than Wata ratings or NRFB value, he's always most interested in this stuff's history, context, and cool factor.

Given Asheville's previous resilience to storms⁠—the town had been called a "climate haven"—and Basinger's archival care with his collection⁠—humidity control and storage "on steel shelves elevated on casters or mounted on walls"—there didn't seem much cause for concern. Basinger even had the trees on his property inspected for structural integrity ahead of the storm.

But then Asheville weathered record rainfall before Helene even arrived, priming the area for the flooding to come. With no evacuation order, Basinger stayed put. "What I did not expect," Basinger said, "were two of the healthiest, largest, heaviest oak trees right in front of my house falling at the same time. I also did not expect them to fully cleave my roof and home right down the middle, front to back." In the face of such damage, and fearing for his own safety, Basinger made the call to leave his home and shelter with a neighbor.

"The room with the majority of my vintage hardware was right below where the trees fell through, and it started filling with rainwater in no time," said Basinger. "After briefly peeking into the collection room and seeing all of the water flowing down from the ceiling onto my 80s and 90s computers and peripherals, I simply sighed and closed the door right behind me. Seeing that absolutely made my heart sink, rapidly followed by numbness at how powerless I was to do much about it."

Collector's value

Miraculously, though, Basinger estimated that "95%" of the collection survived in his most recent video, and he's doing his best to keep things in perspective: "It could have been so much worse, but thankfully it was not," he said the new vlog. "Just looking at my actual neighbors down the road or on the other side of the mountain, unfortunately they don't have a home to even restore. Some people lost their lives⁠—a lot of people did.

"Just extremely grateful for what I've got here: The community, the support, all of it. I'm gonna be fine."

Basinger is running a fundraiser through YouTube for Operation Airdrop, a charity that delivers supplies by air to victims of natural disasters. He also directed viewers toward Blue Ridge Public Radio's page documenting local charities to assist flood victims.

As for assessing his own life and the state of his collection after the disaster, Basinger took it piece by piece once he could safely return. The upper floor of Basinger's home did not collapse, and while the room where he stored most of his collection sustained significant water damage, he was relieved to find that most of his treasures could be salvaged.

"The experience of taking each piece out of there and finding that nearly all of it was intact enough to simply need a wipedown and some drying off in the sun? Yeah, that brought on an unexpected and revitalizing sense of relief," Basinger told me. "I'd spent days wondering if I'd even be able to salvage that room of stuff at all due to how much rain had soaked through, so to find the majority of those cardboard retail packages and computer game boxes intact seriously caught me by surprise."

A great deal of old hardware still took the hit⁠—in his vlog, Basinger stopped to inspect the remains of an absurd Ferrari branded laptop that cost $2,000 in 2005⁠—but he expressed optimism that even some of these pieces can be saved: "There are many retro items in far worse condition that I've brought back to life before."

Basinger also described a newfound perspective on the retro computing hobby. In addition to rethinking having so many valuable items condensed in one place from a practical perspective, Basinger is considering how to ensure this history is best preserved and bringing joy to as many people as possible: "I also think more than ever that there's a need to redistribute a lotta this stuff."

"I'd rather see a good chunk of it spread among other enthusiasts and collectors, hopefully being enjoyed and used more often than I'm able to do myself. I also found myself thinking that if the worst case scenario happened and all of that stuff was irreparably ruined, that there'd be very little of it I'd want to reacquire. I'm not sure what to make of that last part yet, other than some part of me was ready to accept the loss and felt ready to move onto whatever's next. And I'm still ready for that, but I'm also glad that not much of historical significance was lost either."

In addition to improvisational brawling and sneaking, there's another pillar of the Indiana Jones experience MachineGames wants to lean into with the Great Circle: Puzzle solving. Giant stone gears turning improbable ancient machinery, assorted cups and whether a carpenter would be caught dead holding them, I would expect no less from an outing with Dr. Jones. 

In a conversation with PC Gamer global editor-in-chief Phil Savage at Gamescom, MachineGames creative director Jens Andersson explained how these puzzles will challenge us, but ideally not send us to our phones to Google the solution.

"We need these puzzles to be challenging," Andersson explained. "We don't want to dumb them down into this kind of cinematic set-piece: You pull the lever and the old mechanics release, it needs to be proper puzzles."

According to Andersson, the Great Circle will have a large variety of those proper puzzles, which hopefully means we won't get that "ugh, this again" feeling of Half-Life Alyx's hologram minigame or BioShock's blasted pipes. The goal was "finding the balance to make them challenging, but not blocking you," and Andersson even declared that "Players should never get stuck in this game⁠—then we have failed."

It's a thorny balance: I like a good puzzle, and I love when a game surprises me, but we've all run into games that are so obtuse, or bring the pacing to such a grinding halt, that scampering off to a handy guide or Reddit thread just makes sense. To head that behavior off, The Great Circle will have optional in-game hints accessible through Indy's camera.

"We have it as a hint system, so if you do get stuck, we don't want you to go to YouTube to look up the solution, because that's immersion-breaking and bad," said Andersson. "We want to keep you in the game. But we do recognize that skill levels vary. So you can snap a photo of the puzzle, and Indy will help, give you a suggestion, you can keep going if you need more information."

I like the sound of this mechanic, especially since you have to opt into it. The other extreme from being stumped by a game is having your chatty protagonist or their goofy sidekick go all "Hm, a lever, wonder if it'll do anything to that machinery we saw earlier…" the second you enter a room. Thanks for the tip, Dora the Explorer, but I really think I could have handled this one.

I've pretty much gotten used to having my phone or Steam's browser overlay on hand to help with playing games⁠—it's one of those things like save scumming that I don't like in theory, but doesn't bother me all that much when I'm actually playing a game. Still, I'll be interested to see if the Great Circle's in-game hints will help with my immersion. But this all raises another issue: Won't someone please think of the livelihoods of our yeoman videogame guide YouTubers?