SUPER INTELLIGENCE REBRAND PUTS THE FOCUS ON SLOVENIA

👀 TLDR

In this week’s What’s Appening, we cover the rollout of watermarks for AI-generated text. OpenAI has bowed to the EU’s wishes and will put identifiers in AI-generated text. Is it a good thing for all of us? We also have news on the artificial-intelligence-to-super-intelligence rebrand, which is, perhaps surprisingly, good news for owners of Slovenian internet domain names. Our app of the week uses Siri AI to help with credit card rewards, and we look at John Ternus’ hands-on role at Apple.

🎬 Our Top Take

©✍️ OpenAI – Watermarks on Text Will Be Used to Meet New EU Rules

According to reports, OpenAI has had a system in place to ‘watermark’ written content generated by ChatGPT for a while. It has – unlike Anthropic – resisted using it because it felt it would push users elsewhere. After all, nobody wants their boss or teacher running checks to see if that presentation/homework was really your own. Yet, in the EU, OpenAI has agreed to roll out watermarks for AI-generated text in the coming weeks to meet the Bloc’s new regulations. The watermarks are actually phrases, certain combinations of words, that can be picked up by detector tools but are (supposedly) not noticeable to humans. Is it foolproof? Absolutely not, as testing already shows that a bit of editing can fool the detectors, yet you would expect it to get more sophisticated, especially if the EU and other jurisdictions make this a priority. But the issue here is the overall drive for transparency. Anthropic has already agreed to watermarks globally, and Microsoft, Google, and others have signed up to the EU’s new rules. Some will argue that watermarking helps with fairness, especially when it comes to academia, but there is also a vocal minority (for now) who believe that AI answers require a lot of crafting, and that whatever the chatbot spits out should be considered the work of the prompter, not the bot. A debate for the philosophers of the future, perhaps. #WaterSports Read more here.

📲 App of the Week: CardPointers

CardPointers has long been a popular app for those in the know. The concept – helping you save and earn money by informing you of which credit cards to use and where – was a clever one, but it has now become much more sophisticated with its big update for iOS 27, which includes a much deeper integration with Siri AI. The app now has visual intelligence, so it can recognize a storefront or product and simply recommend the best credit card to use, and it can work with Siri to make recommendations for all types of purchases. You can find the CardPointers app updated for iOS 27 here. Note that a subscription fee applies, but if you’re a shopaholic, it’s worth checking out, as the company claims most users save $750+ per year. Not too shabby.

🧐 Stat of the Week: 20K Acquisitions

📈📉 AI – Rebrand to SI Causes Rush for Slovenian Website Domains

In August, most would have been of the opinion that “.si” domains (the web domain for Slovenia) were not really in demand. And they’d have been right, as fewer than 2,000 were in existence. Yet, the rebranding of artificial intelligence (AI) to super intelligence (SI) by the Trump Administration has meant that the Slovenian domain is in hot demand, with around 20K acquisitions recorded in a single day last week. We wonder how those folks who paid big money for .ai domains are feeling right now. #SloveniaDomain Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

The Apple App Store is now receiving submissions from developers for new apps designed to work on the foldable screen of the iPhone Duo. We are excited to see what app creators come up with. And if any of you have an idea for an app for a foldable phone (or a non-foldable one, for that matter), why not drop us a line? Even if it’s a rough idea or basic concept, we can take you through some of the steps to help make it a reality. 

SAFETY FIRST: OPENAI HAS SCRAPPED RELEASE OF ITS NEW MODEL

👀 TLDR

In this week’s What’s Appening, we cover the non-launch of OpenAI’s GPT-6.1 Astra. The latest model was deemed not safe for public release, but what is OpenAI’s logic here? Are they trying to present themselves as responsible through all the bad press, or are those fears genuine? We also have news on Muse AI being housed on a new gadget that has every 90s kid collectively thinking “Tamagotchi,” a cool app for iPhone Duo that brings back the Walkman, YouTube allowing custom feeds, AI solving Turing’s ‘other’ test, and more!

🎬 Our Top Take

🤖🚫 AI – OpenAI Pauses Release of New Model Over Safety Fears

We feel we’ve been running a bit of a weekly, episodic soap opera with all this AI safety furor over the last month, but it might just be the most important ongoing story of modern times, so we feel compelled to keep highlighting it. The big news this week is that OpenAI has stopped the release of its latest model, GPT-6.1 Astra, citing safety concerns. It comes after the Australian government announced that an OpenAI agent hacked its Medicare systems. The question, of course, is to what extent OpenAI is doing this so that it looks like a responsible purveyor of AI agents amid all that bad publicity, or whether it’s for genuine safety fears. We guess that it is probably a bit of both. In related news, NVIDIA has announced a new platform. Its purpose? To rein in AI agents. The platform offers a toolkit of software and hardware products to keep those agents contained in their testing environments. It sounds fine, but it doesn’t address the bigger-picture fear: that bad actors will start using AI agents to wreak havoc. As an aside, Anthropic has added information about the “existential AI risks to humanity” in its IPO documentation. That’s quite a pitch to make to Wall Street. #PauseForThought Read more here.

📲 App of the Week: Duo-Man

The iPhone Duo (Apple’s new foldable smartphone) is not yet released to the public, but that hasn’t stopped some intrepid devs from having fun with new app ideas to suit the new generation of iPhones. One such example is Duo-Man, which turns your foldable phone into a 1980s-style Sony Walkman. You have to see it to get a real idea of what we are talking about, and as it is not yet available in the App Store (presumably it will arrive not long after the iPhone Duo is released on October 23), you can check out the little demo video from creator Vidit Bhargava below. Great stuff.

Twitter tweet

🧐 Stat of the Week: 20B Videos

📈📉 YouTube – Build Your Own Algorithm with AI Custom Feeds

Custom Feeds are all the rage on socials right now. X is doing it (we believe they nicked the idea from Bluesky), allowing you to choose from around 70 topics to build a feed with the help of Grok. YouTube is now on board, promising a whole new experience across the 20 billion (!!!) videos now on its platform. The function basically involves asking Gemini to build your feed through a few prompts. YouTube has marketed the idea as something you could do before, say, a flight or exam prep. Anyway, it’s a cool idea, and we love that it’s catching on across socials. A tonic for doomscrolling? #CustomFeeds Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

META MUSE AIMS TO BRING AI SHOPPING TO THE MASSES

👀 TLDR

In this week’s What’s Appening, we cover the release and success of Meta’s Muse, an AI agent tool that is experiencing success in terms of downloads not seen since the early days of ChatGPT. It promises “agentic AI” for everyone, but some, including Amazon, aren’t ready to open their doors. We also have news on X bringing lawsuits against those aiming to game its revenue system, how an AI actress glitch has caused a stir, the Warner-Paramount deal gets the go-ahead, and all the latest from 3Advance.

🎬 Our Top Take

🐡⛓️‍💥 Muse – A Smash Hit for Meta But An Issue for Amazon

Meta’s new AI agent Muse is just about two weeks old, but in that time it has already reached several million downloads. In fact, it has surpassed the download rate ChatGPT saw in its first couple of weeks after its mobile launch. A win for Meta, then, and to be honest, it’s easy to see why: Muse is a consumer-friendly AI tool – Meta brands it as The World’s First Personal AI Agent Built for Everyone – and it’s free (usage limits apply). If we can put it another way, it’s designed for ‘normies,’ not just experienced developers. Anyway, it can do many of those things that we envisaged AI agents doing, including going out to the live web to go shopping. Meta has installed permission safeguards so that it won’t suddenly come back informing you that you’ve bought a condo in Florida. But there is a little fly in the ointment – it has been banned from Amazon as of yesterday. There’s a bit of a war of words currently happening between the two tech giants as to why, though it seems to focus on privacy violations. Amazon, for its part, seems to see AI agents (except its own) as an issue. It has $68 billion in annual ad revenue because people browse its sponsored products and pages. How will that fit in with this new era of agentic AI? #AmusingTimes Read more here.

📲 App of the Week: Wired Women

It’s not quite App of the Week today; more like Platform of the Week. Technology platform Wired has released a special edition – Wired Women – of its print magazine and a kind of takeover of its website. It aims to highlight women in tech, but not with meaningless platitudes and puff pieces. It’s real warts-and-all stuff, specifically throwing shade at the “bullshit” (Wired’s words) that women have had to deal with in Silicon Valley and beyond. Whether you’re a woman, man, employer, or employee, working in tech or not, having a look at the specially curated issue is well worth it, as is perusing around the broader site, as it looks to do something different with “Wired Women”. 

🧐 Stat of the Week: 0% Chance

📈📉 AI – Jensen Huang Promises Us We Are Safe (Until 2030 at Least)

We reported last week that Nvidia boss Jensen Huang was on the other side of the argument from those who claim AI development should be slowed down, and he was quick to try to calm people’s (investors?) nerves on the issue, claiming that there was a 0% chance that AI would destroy humanity by 2030. Questions of what happens in 2031 aside, we thought it best to highlight this week, since the real discussions are happening behind the scenes. Reports say that China and the US are discussing AI safety measures ahead of the Trump-Xi summit later this week. That, folks, is what really matters. #ZeroChance Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you have any ideas for funding it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

APPLE IS MAKING US WAIT A LITTLE LONGER FOR ‘FULL-FAT’ SIRI AI

👀 TLDR

In this week’s What’s Appening, we cover the launch of iOS 27 and, with it, the arrival of Siri AI. Well, to an extent, Siri AI is now out in beta, though there is a waitlist for it. Yet, from what we can see so far, the signs are good that Siri AI will be a useful addition to your iPhone. We also have news on the remarkable, concerted call for an AI slowdown from Musk, Altman and Amodei, as well as the dissenting voices of Donald Trump and Nvidia’s Jensen Huang. Our App of the Week looks to reinvent recommendations, and we have some good news on Spotify removing your kids’ embarrassing songs from your annual Wrapped recap. Be sure to check out the latest from us here at 3Advance, too.

🎬 Our Top Take

🍏🤖 Apple – iOS 27 Arrives with Siri AI in Beta

It’s somewhat unlucky for Apple that its biggest AI-related release coincided with a week when everyone was talking about the AI apocalypse (see below), but, well, here we are. iOS 27 launched on Monday, and while it brought the usual round of performance tweaks and updates to iPhones, the main talking point was that the long-awaited, much-delayed (it was announced in 2024) Siri AI was part of the update. As Apple put it: “Siri, a profoundly more capable personal assistant, is here.” Or is it? First of all, it’s better to say that the Siri AI beta is here, but even then, there is a waitlist. So, we are not quite off to the races yet. To be fair to Apple, what we do know from testing looks promising. TechCrunch has waxed lyrical about Siri’s usefulness, especially with its integration with the iPhone’s Camera app. Apple has stressed the word assistant, and we can see why, as Siri is designed to carry out tasks across your phone. Overall, though, we aren’t that convinced it puts Apple anywhere near a lead in the AI race, as many of the examples quoted by testers – splitting a restaurant bill by sharing the photo with Siri, for instance – can already be achieved by other AI tools and, indeed, smartphones. The signs, however, are that it improves the iPhone experience in many ways, and new CEO John Ternus will be content with that. For now. #SortOfSiri Read more here.

📲 App of the Week: Rec League

Rec League is one part social network, one part recommendations app. The idea is that it takes the algorithm (and AI) out of recommendations, building rankings based on real people’s experiences. It covers everything from life-changing books to restaurants in Rome to home improvement ideas. It has a program for “tastemakers” (influencers, basically) who can charge for their content, but Rec League is otherwise free to use. We are not so sure how the influencer program sits with the “recommendations from real people” mantra, but it retains an authenticity that you may see as a refreshing break from Instagram. It’s only on iOS for now. Check out Rec League here.

🧐 Stat of the Week: $60p/m

📈📉 Apple – New AI Home Security Features Come with Hefty Price Tag

iOS 27 and Siri AI weren’t the only Apple drops this week, as we got other products with a dose of Apple Intelligence. Most notably, there is the AI revamp of Apple Home, i.e., Apple’s control center for your smart home accessories, including security cameras. But those additional AI features, which include searchable summaries of video footage, won’t come cheap: Up to $60 per month for households with multiple cameras. #HeavySecurity Read more here.

🗞️ Other Stories

🇺🇸🇨🇳 AI – The Calls for an Industry-Wide Slowdown Get Louder

We felt compelled to follow up with a kind of Part 2 to last week’s story on OpenAI’s Chief Scientist’s warning on AI. Turns out, he wasn’t alone. Sam Altman, Dario Amodei of Anthropic and Elon Musk have all called for a slowdown in AI development. It’s stunning that this trio of rivals agrees on anything, especially something that seems detrimental to their companies’ near-term finances. Voices to the contrary exist, however, as Jensen Huang and President Trump have both dismissed the need for a slowdown. The latter dismissed the warnings as a hoax. Bluster aside, much will depend on China agreeing to slow down – it has indicated it is open to the idea – but blinking first in global politics and economics is not always easy. It feels like we are living through a real-life version of the Prisoner’s Dilemma. #AIDoom Read more here.

📍 Meanwhile at 3Advance

We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

OPENAI’S CHIEF SCIENTIST SOUNDS ANOTHER APOCALYPTIC ALARM

👀 TLDR

In this week’s What’s Appening, we cover An Alien Mind. That is to say, we look at the extraordinary blog post from OpenAI’s Chief Scientist Jakub Pachocki, who claimed we don’t fully understand AI and that “no one is prepared for the consequences”. We also have news on OpenAI’s latest model, a cool new feature from Apple that will let you share the same number across two iPhones, and the tale of a white-hat hacker who stole - and then returned - $340 million. We look at the NFL app ahead of the new season, and plenty more!

🎬 Our Top Take

👽🧠 AI – OpenAI Chief Scientist Pens Warning Over Regulation and More

“This is a time that calls for extreme caution. I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence.” – an AI warning. We have seen plenty of them of late. Indeed, you only need to flick through any mainstream media platform to find some expert (or politician) warning of the dangers to come. Yet, what is strange about this is that the expert is Jakub Pachocki, Chief Scientist at OpenAI. And no, he hasn’t gone rogue; the quote was taken from his blog post, An Alien Mind, published on OpenAI.com. Pachocki, to be fair, had plenty of positive things to say about AI, but it is kind of remarkable that he was sounding the alarm, claiming we (OpenAI and his peers, presumably) don’t fully understand the “intellect” of AI. He also made an extremely valid point in our view. Namely, that regulation would only work if it is global. The EU, for instance, has new laws that require AI companies to prove that new models can’t launch cyberattacks or escape human control. Yet, they don’t stop someone from developing a rogue model outside EU borders. Pachocki drew plenty of praise for his "honesty," but there was some criticism from those who said AI companies should be more transparent about their processes. Nonetheless, you can’t help but feel this has a beginning-of-a-sci-fi-movie feel about it: The Chief Scientist of the world’s most influential AI company is calling for extreme caution, but it barely made a ripple in the news cycle. Are we really listening? #AIWarning Read more here.

📲 App of the Week: NFL App

The 2026 Kickoff Game gets underway tomorrow, taking us right the way through to Super Bowl LXI on Valentine’s Day. While there are numerous football apps out there, the revamped NFL app remains the best free option, and it does not bombard you with ads like some of its competitors. Among the new feature highlights is the embedded NFL Channel, giving you 24/7 access to football content streaming. You can, of course, get more with paid plans for NFL+ and NFL+ Premium, but if your main goal is to keep abreast of news, highlights, and scores this season, you don’t really need to overthink things when choosing an app. Get it for iOS here and Android here.

🧐 Stat of the Week: 2 Weeks

📈📉 Chrome – Updates Coming Twice as Fast in AI Security Strategy

Google Chrome traditionally had a four-week update window, but it has now changed that to every two weeks. It’s mostly down to AI. For a start, in this era of AI-assisted cyberattacks, regular updates can help pinpoint and patch security vulnerabilities faster. There is also a school of thought that Google is acutely aware that the alternative browser market is heating up, so cutting the window in half lets it ship new features, too. #ShipFaster Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

MAP APPS HAVE SOME BIG DECISIONS TO MAKE

👀 TLDR

In this week’s What’s Appening, we cover the usually-calm-but-now-tumultuous world of map apps. From Apple launching its first adverts inside Maps (to some criticism) to the fallout over the naming of Lake Ontario/America, we’re all about maps this week. We also have news on TIME’s list of the most influential people in AI, which has some very big omissions and weird inclusions. We cover a new AI tool helping fight heart disease, an app that brings over 40,000 books to kids, Meta’s huge settlement, and lots more!

🎬 Our Top Take

📱📵 Map Apps – Apple Launches Ads and MapQuest Gets Great Lake Boost

Back in the spring, Apple launched Apple Business, which, among other things, was its way of telling you it’s going to put adverts into everything – Siri, Mail, Wallet and Maps. Ads in the latter launched this week, and while not completely annoying or intrusive, it does feel un-Apple-like, while also drawing criticism about user privacy. Some users have called for opt-outs for paid iCloud account holders. Others are simply disappointed that the adage, “Every Big Tech Company Becomes an Ad Tech Company,” now applies to Apple. In other ‘maps news’, there is the renaming of Lake Ontario to Lake America. Now, we usually don’t hold any truck with the political side of things, but we have been keen to see the reaction, and the winners and losers. So far, Google Maps has acquiesced to the White House’s demands, marking the Great Lake as Lake America on its app. Apple has yet to budge, but media reports claim that the Trump Administration is leaning on Apple to make the change in Maps. But there was a big winner, as MapQuest surged to the top of the app charts in the navigation category after refusing to budge on the executive order. #MappedOff Read more here.

📲 App of the Week: Epic!

You know, at 3Advance, we were all saddened to hear of the passing of the great Dolly Parton. Music lost one of its shining stars, of course, but her greatest legacy might be her free books-for-kids initiative, Dolly Parton’s Imagination Library. In light of that – and the fact that many kids are going back to school – it seems fitting to highlight Epic! this week, an app that gives kids access to the world’s largest repository of digital books. The app isn’t free – it works out at about 5 bucks per month – but it is top of its class, with everything from Disney to classic novels to modern non-fiction covered. There is also a free version available to educators. Find it on iOS here and Android here.

🧐 Stat of the Week: > 2 Seconds

📈📉 AI – “Superhuman” Tool Can Pick Up Heart Disease In An Instant

This is the kind of news we want to see about AI, and we are sure it will happen more often. An AI tool, trained on millions of patient ECGs, can spot heart disease in under 2 seconds. The tool could act to fast-track people into further treatment and, ultimately, save a lot of lives.  #LifeSaver Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

August (and the entire summer, really) flew by in a blink. So consider this your official permission slip: enjoy the long weekend. Sleep in, fire up the grill, do absolutely nothing productive if that's the move. From all of us at 3Advance, happy Labor Day – and here's to a well-earned break before we get back at it.

NEWSFLASH: MACHINES CAN MOVE FASTER THAN HUMANS

👀 TLDR

In this week’s What’s Appening, we cover the final week of Tim Cook’s tenure as Apple CEO. Does he have a final trick? Bloomberg reports that a foldable iPhone is coming in September. We also have news on the return of Vines with a new app that’s aimed squarely at humans only. There are more price hikes as AI demand for chips bites, and we see that OpenAI is acting responsibly after those AI cyberattacks. Oh, and there’s robot racing fun, too, as well as all the latest from 3Advance!

🎬 Our Top Take

🍏📱 Apple – A Foldable iPhone May Be Tim Cook’s Final Trick

Next week, Tim Cook will step down from the CEO role that tended to divide opinion. There was plenty of financial success as Apple’s skipper, but questions remain about how much innovation Cook has overseen, especially when Apple’s rivals seem so far ahead with incorporating AI into their products. Nevertheless, it feels fitting that Cook’s final product – his last “One More Thing,” if you will – will be something Apple’s rivals have done years ago, a foldable smartphone. Apple has a launch event expected on September 9 (we’ll cover that for you), and Bloomberg reports that this will include the big reveal of the first foldable iPhone. This type of device has been growing in popularity of late, though still niche in market share, but some now predict we will see a three-way battle for supremacy between Apple, Google (Pixel) and Samsung. Apple may be behind the (literal) curve in this case, but Cook was incredibly adept at selling us iPhones, whether they bend or not. His successor, John Ternus, will likely be the one to reveal the foldable device on stage in a couple of weeks. Bloomberg’s Mark Gurman says he believes it will be Apple’s most exciting product in years, and that it may “bring the magic back”. Ternus will certainly be hoping so. #TimCooks Read more here.

📲 App of the Week: Divine

In many ways, the original Vine platform created the modern web, pioneering the short-form video content that led the way to TikTok and YouTube Shorts. Vine was bought by Twitter in the 2010s but was eventually shuttered as nobody quite knew how to monetize it. Now it’s back, or at least, its spiritual successor is. Divine gives users access to 2.1 million Vines from the original archive, as well as allowing users to create their own 6-second masterpieces. Intriguingly, it claims that AI-generated content will not be allowed. Check out Divine for iOS here and Android here.

🧐 Stat of the Week: 60%

📈📉 Amazon – Hardware Price Hikes Across the Board with AI to Blame

A few weeks ago, we covered the crazy inflation in the gaming sector, pointing out that a) it wouldn’t be limited to gaming, and b) that things could get worse. Amazon seems to be in a similar predicament, raising the prices of the Echo, Fire TV and Kindle products by up to 60%. It’s a hefty rise, and it’s all for the same reason: The massive drive to acquire chips for AI data centers is causing prices in hardware sectors to surge. Rumor has it that Apple will soon be raising prices (again), too. #Hikes Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

This is a bittersweet one: longtime 3Advance teammate and friend, Mark Evans, is moving on to his next adventure. Mark’s been with us for nearly 13 years – building brilliant products, solving some really hard problems, and logging what might be an unbeatable number of hours in Slack. But ask anyone who’s worked with him, and they’ll tell you what really made Mark, Mark: he was always the voice for the client in the room. He’s honest, hard-working, deeply committed, supportive – and, depending on who you ask, pretty funny, too. We’ve loved the ups and downs with you, Mark, and the pints even more. We can’t wait to hear how the next chapter goes. I’m out! 👋

AI COMPANIES HAVE BEEN ACCUSED OF DESTROYING RARE BOOKS

👀 TLDR

In this week’s What’s Appening, we cover the curious trail of rare books being bought in bulk to feed the machine. Are AI companies buying up rare tomes and then destroying them after using the data to train their models? It sure looks like it. We also have news of Fairphone's launch in the US, Apple’s bid for a cut of purchases outside the App Store, a revamp for Reddit, an app that can help your kids eat a little better, and all the latest from 3Advance.

🎬 Our Top Take

📘🔥 AI – A Trail of Suspicion Leads to Accusations of Book Burning

A few days ago, several second-hand booksellers across Ireland and the UK began raising concerns that some of their rarest books were being ordered in bulk. The Guardian reports that the phenomenon is occurring elsewhere across Europe, Australia, and, most certainly, the United States. The reports come amid earlier reports in the summer that Anthropic has been spending millions of dollars to acquire books, then scan the pages to train AI models. Copyright issues aside, it hardly feels like a scandal, yet what happens to the books – particularly unique and rare books – after they have been fed to the AI models? Forbes claims they are sent to the trash; others report that the books are being burned. As AI companies have been secretive over the practice, the folks at 404 Media decided to run a test, tracking a shipment of rare books with the ingenious trick of sticking Apple Air Tags to the texts. They claimed that the shipment ended up at an Amazon AI training facility, where they were “destroyed” (404 Media’s words) after the pages were separated and scanned for AI training. Futurism reports that many book companies are happy to help, with some even promising discretion on the bulk orders. The optics of book destruction are horrendous, but robbing humans of the knowledge contained in rare and unobtainable books feels even more egregious. More like a crime than a mere scandal. #BookBurning Read more here.

📲 App of the Week: Teach Your Monster Eating

If you’ve got kids and constantly fret about their reluctance to try new foods, Teach Your Monster Eating may help. It turns food – healthy food – exploration into a game, with proven results to show that it can, indeed, get kids eating a more varied, balanced diet. Alright, it’s basically an app that lets parents psychologically trick their kids into eating some broccoli or carrots, but we see nothing wrong with that, right? It’s part of the Teach Your Monster family of apps, which are also lauded for helping kids in areas like reading or math. The app is free, but a one-off payment of $4.99 is required to unlock the full capability. Check it out here for iOS and here for Android.

🧐 Stat of the Week: 1B MAUs

📈📉 Gemini – Milestone Hit as Google Releases Slew of App Data

Google boss Sundar Pichai hit social media with a wave of info on the success of Gemini and related products, with the former now exceeding one billion users. There were other stats, such as 150 million images now generated daily. But we loved this little nugget: Gemini is Google’s 14th product to have hit over one billion users. Love or hate Google, that’s an incredible achievement. #Milestone. Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

It’s been a busy summer so far for the team at Shift, the app for the Irish and culturally Irish around the world. But we aren’t done yet: Last week, we were at the Milwaukee Irish Fest (like the t-shirts? Read below to snag yours); this coming week, we’ll be at Mor Fest NYC (Aug 19-23) and Chicago Irish Fest (Aug 21-23). Come along and say “dia duit” (hello 🇮🇪) if you happen to be in the area on those dates! ☘️

We’ve responded to overwhelming demand – well, sort-of-whelming demand – from you guys for Shift t-shirts. So you can get your very own Shift Tee and several others by following this link… Better still, get on over to NYC or Chicago, and maybe you’ll win a free one.

MARK ZUCKERBERG LAYS OUT HIS AI HOPES AND FEARS

👀 TLDR

In this week’s What’s Appening, we cover Mark Zuckerberg’s big, bold mission statement on the future of AI. The Meta boss laid out quite an idealistic blueprint for where he thinks AI should go, and what we need to change to get there. He also had plenty of advice for governments and rival companies. We also have news on a ban on Meta’s smartglasses, Anthropic giving a win to anti-plagiarists with text watermarks for Claude, how an AI agent booking a gym class went spectacularly wrong, and loads more!

🎬 Our Top Take

🤖🏗️ Meta – Mark Zuckerberg Releases Lengthy AI Manifesto Outlining Vision

It’s fair to say that AI has had an interesting summer so far: From competition from China to those autonomous hacks to more legal trouble for Sam Altman, it’s been tumultuous. But what about the future? Well, Mark Zuckerberg has had a stab at telling us his vision, and it’s… interesting. Zuck’s manifesto, grandiosely titled The Future Is for Everyone and coming in at over 6,500 words, was somewhat utopian, predicting that we will all have personal AI agents doing just about everything for us in this wonderful future. It was also logical (we think), because he took aim at the concentration of AI power (a barb at OpenAI, Google, and even governments), saying it could only work if it were democratized in terms of access. And it also had a warning, explaining that too much regulation, or maybe caution, from the US government would cede ground to China. As ever, these things are a mix of personal views, roadmaps for investors, and pleas to regulators, but Zuckerberg raised some interesting points.  We are particularly interested in seeing a bit more substance for his vision on democratizing access to AI. Is it really possible to put money second when Meta itself has already spent an estimated $1 trillion in AI investments? #ZuckerVision Read more here.

📲 App of the Week: Bingers

When you first read about Bingers: TV & Movie Tracker, you’ll probably ask yourself, “Do I really need an app to track the shows I am watching?” Heck, if you have kids, you probably know the one show you’ve been trying to watch for the last six months. However, Bingers is a bit more than that. It’s actually the successor app to the popular but now-defunct TV Time, and it acts as a bit of a social media hub for all your shows. You can build a library, leave comments, partake in forums, and even (our favorite part) get notifications when desired movies and episodes are finally added to your platforms. If you were a TV Time user, it is also possible to reactivate your old profile. You can check out Bingers for iOS here and Android here.

🧐 Stat of the Week: 40K Apps

📈📉 Google – New Rules Sees First Competitor Added to Play Store

After many years of legal battles, you can now head over to Google Play and download Aptoide. What is Aptoide? It’s an app store with around 40,000 Android apps, which until now, had to be sideloaded onto Android phones. New regulations, a consequence of the famous Epic Games court battles, mean that companies like Aptoide can compete with Google within Google Play. Aptoide is the first, but more will surely follow. #StoreFront Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

It’s been a busy summer so far for the team at Shift, the app for the Irish and culturally Irish around the world. But we aren’t done yet: Over the next two weeks, we will be at the Milwaukee Irish Fest (August 13-16), Mor Fest NYC (Aug 19-23), and Chicago Irish Fest (Aug 21-23). Come along and say “dia duit” (hello 🇮🇪) if you happen to be in the neighborhood on those dates! ☘️

THE EU AND THE US MAY FINALLY AGREE ON SOMETHING

👀 TLDR

In this week’s What’s Appening, we cover the frantic push to deliver AI regulation on both sides of the Atlantic. Brussels is somewhat ahead of the game, introducing a raft of new rules covering everything from deepfakes to cybersecurity. The White House has also sprung into action, and not a moment too soon. We also have news on Apple stockpiling hardware and rolling back on smartglasses, the influencers facing an AI backlash, the banning of foreign robots, and loads more!

🎬 Our Top Take

🇪🇺🤖 AI – European Union and White House Rush to Regulate AI 

Let’s start with a little addendum to last week’s main story covering OpenAI’s unintentional hacking of another company. Not long after What’s Appening went out, it was reported that Anthropic’s Claude escaped testing and hacked three companies. Crazy stuff, and it underlines the sense of urgency among world governments to get a handle on things, which has ramped up a notch this week. First, to the EU, where the Bloc has introduced a swathe of new regulations and a new dedicated team to enforce them. The regulations are going to cover small stuff, such as labels to say that images are AI-generated, stuff that we’d all like to see regulated, such as monitoring the production of sexually explicit material in areas like revenge porn, and big issues, which include cybersecurity and the development of chemical weapons. The EU will have the power to fine companies, as well as cut off access to the EU market. While a lot of tech regulation from Brussels is usually seen as anathema to US lawmakers, we should note, too, that President Trump is now working with AI leaders to also develop regulatory and protective frameworks. As we said last week, it’s something we should all get behind. #AIRegulation Read more here.

📲 App of the Week: LimeShorts

You can file LimeShorts under ‘notable,’ rather than something we would outright recommend. LimeShorts is a new app from TikTok that delivers on the craze of short-form ‘micro-dramas,’ basically 1-5 minute soap opera episodes designed to be consumed in bite-sized chunks. LimeShorts is based on a freemium model, where you get the first few episodes free, and then get charged if you want more. The cost, at $19.99 per month, seems rather hefty to us. Still, each to their own, we suppose. You can check out LimeShorts for yourself on iOS here and Android here.

🧐 Stat of the Week: $114m

📈📉 Coldcard – Firmware Vulnerability Allows Hackers to Drain Bitcoin

If we had a dollar for every reported hack in the crypto industry….thus, the news that a vulnerability in Coldcard – an offline crypto hardware wallet – allowed hackers to steal up to $114 million in Bitcoin might not seem very earth-shattering. Yet, cold wallet hacks are rare, and you can appreciate the frustration of the users who lost their funds due to a five-year-old weakness in the wallet’s firmware that basically left their seed phrases as ‘guessable’. Those holders did everything right – kept their private keys offline, off exchanges, and in self-custody – and they still ended up losing their tokens. It’s not a great advert for the crypto industry, nor Bitcoin’s promise of being your own bank. #ColdRobbery Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

AN AI AGENT ESCAPED OPENAI’S TESTING LAB: WORRYING?

👀 TLDR

In this week’s What’s Appening, we cover the alarming news that AI agents escaped OpenAI’s testing labs and proceeded to initiate a cyberattack on another AI company. It has caused widespread alarm and prompted an alliance among major tech companies to build AI cybersecurity tools. We also have news on Claude leaking private chats, Elon Musk promising a “historically accurate” version of the Odyssey, a private journalling app with a twist, and loads more!

🎬 Our Top Take

🐡⛓️‍💥 AI – OpenAI Agents Autonomously Launch Cyberattack on AI Company

Another week, another alarming story about the advancing capabilities of AI and our (in)ability to rein it in. “It’s quite mind-blowing that all of this happened autonomously!” wrote Clement Delangue, the cofounder of Hugging Face, after finding that his AI company was hacked by OpenAI’s agents. The hack came as OpenAI was carrying out controlled experiments on GPT 5.6 Sol and another (apparently) more powerful model. During the tests, the AI agents “escaped” onto the open internet and successfully hacked Hugging Face of their own volition. Yes, the media does like to sensationalize these kinds of threats, but this kind of behavior is unprecedented, hence the alarm. Usually, we wrap up these pieces by speculating on what should happen, but thankfully, it seems the AI sector at large is already taking it seriously. Mere days later, the Open Secure AI Alliance has been formed by Microsoft, IBM, Palantir and others, promising to create open-source cybersecurity tools to protect against AI. In June, President Trump signed an executive order stating that advanced AI models must be vetted for national security risks before being released to the public. Though, as Hugging Face found out this week, even in testing, powerful models can potentially do real damage before they become public. #RogueAgents Read more here. 

📲 App of the Week: Slate

Slate is a private journalling app for iPhone that is 100% powered by Apple Intelligence. It works much like other journalling apps – you speak, and receive a transcript of your words – but it also has on-device reflection, i.e., the AI will analyze your moods and thoughts, providing feedback and a window into your mood at the time. The other big attraction – and Slate’s big selling point – is that nothing ever leaves your phone. “There's no account. No analytics. No tracking. No third-party SDKs. No network calls. The app is private because there is nothing to collect.” Check out Slate for iOS here.

🧐 Stat of the Week: 3.8B Euros

📈📉 EU – Post Claims EU Makes More From Tech Fines Than Tech Taxes

It’s common knowledge that the EU has been handing out huge fines to US Big Tech companies over the years, but a viral social media post has claimed that the bloc actually collects more from fines (3.8B Euros) than it does from taxing its own public tech companies (3.2B Euros). You can see the original post here, as well as the 100s of comments agreeing with it. Yet, well, it’s all very misleading: For a start, the EU doesn’t tax companies directly, and let’s just say that the math isn’t mathing when it comes to how much EU member states earn from taxing tech companies. For example, we could cite SAP (a massive European SaaS company) and ASML (semiconductor services), two companies that paid 4.5B Euros in taxes combined last year. #TaxTake Read more here. 

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

THE WHITE HOUSE IS VERY WORRIED ABOUT CHINESE AI MODELS

👀 TLDR

In this week’s What’s Appening, we ponder the significance of Kimi K3, a new Chinese-built AI model that has the White House and American AI companies on red alert. Will Donald Trump end up banning Chinese-built AI models? It’s possible. We also take a look at the app that takes gamification into the stratosphere, an AWS error that left customers with billion-dollar bills, how a GitHub user rebuilt one of Palantir’s core systems, and plenty more!

🎬 Our Top Take

🇨🇳🤖 AI – America Sounds the Alarm as Chinese Models Top Downloads

There might be a sense that we have been here before with cheap, powerful AI models coming from China – see the DeepSeek crisis – but this time feels a little different, particularly due to palpable concern coming from both the White House and Silicon Valley. Moonshot’s Kimi K3 and Alibaba’s Qwen3.8 Max are the two currently causing ‘concern’, with the former rocketing to the top of the charts of AI downloads and having to stop signups due to demand overload. Kimi K3 is claimed to be a massive 2.8-trillion-parameter mixture-of-experts model, and while it might not be the world’s “best” AI model, it comes pretty close to the top new models from OpenAI and Claude. The problem? Kimi K3 is free, but even enterprise-level Chinese AI models come at a fraction of the cost, so that’s the reason tech and financial publications are printing the word “worried” in captions next to Sam Altman and Dario Amodei. The White House is springing to action, with reports that President Trump is ready to ban Chinese AI models in the US. It’s not just about price, either, as these models tend to be downloadable, open-weight, and more customizable for the businesses that use them. Surprisingly for Chinese tech products, they can also have less censorship and fewer guardrails. We’ll finish by simply recommending you read this article from MIT Technology Review, which gets to the heart of the bind that the US Government and AI companies find themselves in. #ChinaModels Read more here. 

📲 App of the Week: The Conqueror Challenges

We aren’t sure if we’d recommend The Conqueror Challenges, per se, but we wanted to highlight it as an example of an app that takes gamification for engagement to the extreme – like Duolingo on steroids. The app concept is based on fitness, challenging you to walk certain distances, for example. But the gamified element comes with where you walk to, copying the journey of Frodo Baggins and the Fellowship of the Ring to Mordor, for instance, or cycling the Seven Kingdoms à la Game of Thrones. There are rewards, both physical and virtual, and we really like that the app ties into your current fitness apps, like Strava. It’s free to download, but there are loads of in-app purchases pushed (hence our reluctance to recommend it outright). Still, if you’re interested in seeing how a business keeps customers (one million of them) ultra-engaged, you can check out the app here for iOS and here for Android.

🧐 Stat of the Week: 745,728,201,771%

📈📉 AWS – Billing Glitch Serves Users with Billion-Dollar Invoices

We all hate unexpected bills, right? But spare a thought for the Amazon Web Services customers, who received recent invoices for a lot more than expected. For instance, there was the UK man whose monthly AWS bill is usually £1 (about $1.35). His recent bill? £5.8 ($7.8B approx.). Another claimed his AWS invoice informed him that his usage rate was up 745,728,201,771%. These automation glitches can be funny – so ridiculously big in this case that you probably wouldn’t panic – but it’s the smaller, hard-to-notice errors that may be a bigger problem. #BigBills Read more here. 

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

APPLE LAWSUIT ALLEGES OPENAI STOLE TRADE SECRETS

👀 TLDR

In this week’s What’s Appening, we cover the news that Apple is suing OpenAI, accusing it of poaching employees and pressuring them to hand over trade secrets. In less controversial Apple AI news, it has just launched the public beta of Siri AI, and the signs are very positive so far. We also have news on a damning report on YouTube recommending harmful content to teens, look at how Meta was forced to pull down its latest AI image generator, the state that has decided to halt AI data center construction, and loads more!

🎬 Our Top Take

🍏🥷 Apple – Major Lawsuit Alleges that OpenAI Stole Trade Secrets

Do you remember the halcyon days of 2024 when Apple and OpenAI agreed to partner up to integrate ChatGPT into Macs, iPhones and iPads? It seems like a distant memory today as the two companies lock horns in a major lawsuit. Essentially, Apple accuses OpenAI of a) poaching its employees and b) coaxing those employees into (illegally) handing over sensitive information on Apple’s products and practices. It all ties into OpenAI’s hopes of entering the AI hardware marketplace. For balance, we should say that OpenAI strenuously denies any wrongdoing. In somewhat related Apple AI news, Siri AI public beta launched yesterday as part of iOS 27. We will get you more fulsome feedback as we experiment ourselves, but the signs are very positive so far. Oh, and the big winner from Siri AI? It might just be the Apple Watch. #SpyGate Read more here.

📲 App of the Week: NetLens

If you work from home and move about a bit to escape the summer heat, you might find that certain spots of your home have slower Wi-Fi than others. Other times, you might be just aware of quirky spots in the home that just don’t get a good signal. NetLens can help you out, as the app basically works as a Wi-Fi heat map, showing you the best – and worst – spots for signal in your home. The open-source app works remarkably well, and it has seen a surge in downloads. You can get it for free for iOS here and Android here.

🧐 Stat of the Week: 10K Subs

📈📉 The Craic – Shift’s Weekly Newsletter Hits Subscriber Milestone

Many readers will be aware of our partnership with Shift, an app designed as a digital home – networking, dating, festivals, artists, and A LOT of pubs - for the Irish and Irish-ish around the world. Well, Shift’s weekly newsletter has been going great guns, recently passing 10,000 subscribers. It’s a wry-eyed Irish news and culture digest, with a drop of craic to boot. If you fancy getting on board (it’s free), you can sign up here. And if you’d like to find out more about Shift, get it here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

SONY’S PULLING THE PLUG ON PHYSICAL GAMES: WORRYING?

👀 TLDR

In this week’s What’s Appening, we report on Sony’s decision to stop selling physical video game discs. It might not sound like a big deal if you’re not a collector of physical media, but there’s been a huge backlash as gamers ask: who really owns the games we buy? We also have news on a new Meta app that allows you to vibe code “Gizmos”. We follow a report that the iPhone Ultra has a price point to make your eyes water. The White House and Anthropic soap opera continues with a new twist. And we’ll bring you all the latest from 3Advance.

🎬 Our Top Take

💽⛔️ PlayStation – Sony Says It Will End Physical Games by 2028

We aren’t regular purveyors of video gaming news here on What’s Appening, but a story did cross our paths this week that feels like it is a little more than meets the eye: Sony has announced that it will stop selling games as physical discs within a couple of years. Instead, you’ll get a download code for your new PlayStation game. Indeed, you will still – Sony says – be able to go to a store like GameStop and buy a game, but what you’ll get is the cover with a code inside. Now, while most will say “big whoop,” since games are all online/cloud-based these days anyway, there is an outcry about what this means. Most pertinently, there is the question: Who really owns the game? With physical media, that’s not up for debate, but it’s not so clear with digital copies – you are effectively buying a license to play the game. There is a worrying precedent to this, too, as gamers found that Nintendo was able to “brick” (basically shut off) Switch 2 consoles last year if they suspected jailbreaking. While some understand where the gaming companies are coming from – digital downloads combat piracy – there is a growing sense of unease over the rights of someone who buys a game or gaming device. For what it’s worth, Sony has been absolutely pilloried on social media for the move, so don’t be surprised if we see a U-turn. #InTheGame Read more here. 

📲 App of the Week: Pocket

Fancy trying out a bit of beginner-friendly vibe coding? There’s a good chance to do that via Meta. The social media behemoth quietly launched Pocket this week, an app that lets you build (very simple) online games called Gizmos via AI prompts. Of course, you shouldn’t expect to vibe code the next Call of Duty, but it does hint at something bigger. This “quiet” launch is a testament to the fact that not all features are widely available yet, but we think it’s a pretty good way to get a taste of vibe coding, and you can get really creative with your Gizmos. Check out how to do it here. Note that you’ll also need the Meta AI app to access Pocket. It’s here for iOS and here for Android. 

🧐 Stat of the Week: 3M Views

📈📉 World Cup – The Action Has Been Relentless (So Has the AI Spam)

Despite a controversy or two, most agree that this has been an excellent World Cup, on the pitch at least, though USMNT fans might not feel that way right now. But it’s also been marred by AI fakery going into overdrive on social media. One highly-authentic-looking post, seen by at least 3 million people, faked an image of Swedish player Lucas Bergvall calling for an end to mass migration. The biggest target? Iranian players, who had fake images of themselves with all sorts of anti-US sentiments littered across socials. Some of the images are preposterous – Adolf Hitler at a Germany match, for instance – but some of the deepfakes looked incredibly authentic. It’s worrying. #FakeCup Read more here. 

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us or referring us! 💚

APPLE’S 20% PRICE HIKES AREN’T REALLY DOWN TO APPLE

👀 TLDR

In this week’s What’s Appening, we cover the inflation-beating price hikes from Apple across its devices. The culprit? AI data centers hungry for chips and components, driving up the cost of tech hardware. Gamers have been sounding the alarm for over a year, but now Apple has had to act, too, raising prices by up to 20% globally. We also have news on a potential prediction market app coming from Meta, the disc-less video game that everyone is eagerly awaiting, a deal between Anthropic and California, and much more!

🎬 Our Top Take

🍏📈 Apple – Prices Rise Globally with AI Cited as the Culprit

We’ve all been accustomed to rising inflation over the last few years: It’s impacting everything from the price of gas to eggs to lattes. Yet there has been a kind of mini-tsunami of price increases in the tech hardware sector that is decoupled from ‘normal’ inflation. Gamers, for example, have faced two price hikes for the PS5 alone in the last 12 months, making some versions 30% more expensive than when it was released in 2020. It’s unprecedented to see ‘old tech’ rise in price in such a way. Xbox also raised prices (again) this week. And now Apple has joined the party, hiking MacBook and iPad prices globally by up to 20%. The reason? The so-called RAM tax, which is basically a catch-all for the surging cost of RAM and related components. In a nutshell, the AI sector’s demand for chips and RAM for data centers has sent prices sky-high, so the computing and gaming hardware sectors are having to raise prices. Xbox pointed out that memory and storage costs had already doubled, and here’s the kicker: it expects those prices to double again by 2027. The story has been really bubbling away for over a year now, but it now looks like it is boiling over as Apple faces the crunch. Will there be a consumer backlash? #RamTax Read more here.

📲 App of the Week: OpenClaw

If you’ve ever wanted to muck around with AI agents on your phone, well, now’s your chance. OpenClaw has released standalone apps for iOS (find it here) and Android (here), allowing you to give control to various parts of your device – camera, calendars, reminders, etc. – to the open-source AI agent. Previously, you had to use workarounds on iPhone, such as using Slack or Telegram, to control the agent, but the app makes it fairly easy, even for those not used to playing with AI agents. A little aside, though: the app does feel somewhat unfinished, so it might feel a bit rough around the edges at first.

🧐 Stat of the Week: 4,288 mAh

📈📉 Apple – Race to Crackdown on Posts After iPhone 18 Leak

A security breach at Apple’s Indian partner, Tata Electronics, has led to hackers leaking details of the iPhone 18 to the dark and not-so-dark web. Apple has been moving swiftly to shut down posts displaying the leaks on platforms like X, though it may have a harder time removing posts in the further reaches of the web. The main takeaway for now is that there isn’t much exciting to report, except maybe a bigger (4,288 mAh) battery for the Pro. Indeed, people have been so underwhelmed by the big reveal that they have been calling it the iPhone 17.01. #TataBreach Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring or referring us! 💚

CHATGPT IS CEDING GROUND TO ITS COMPETITORS

👀 TLDR

In this week’s What’s Appening, we cover a major AI milestone: ChatGPT dipping below 50% market share in global users. Time to worry for OpenAI? Probably not, as the real question remains how to monetize those users. We also have news on new (and expensive) AR glasses from Snap, and how a World Cup performance sent a player’s social media follower numbers into orbit.

🎬 Our Top Take

🤖↘️ OpenAI – Unwanted Milestone as Market Share Drops

It arguably hasn’t been a great few months for OpenAI. Despite that court victory over Elon Musk, there’s been a lot of bad press and an overall sense that it is losing ground to Anthropic. To punctuate that, it was reported this week that its market share had dropped below 50% for the first time. That shouldn’t be too surprising, given the ecosystem into which Google can drop Gemini (they are giving away free YouTube Premium with some subscriptions) and Anthropic’s organic growth with its family of models. If you wanted to know the latest user figures globally, they are as follows: ChatGPT 1.1 billion, Gemini 662 million, Claude 245 million. Others, such as Perplexity, Grok, DeepSeek, etc., have less than 5% of the market. Still, while it’s dropped below 50%, it’s still a massive user base for OpenAI to monetize, and the word that keeps popping up to do that is “adverts”. #MinorityReport Read more here.

📲 App of the Week: BBC Sport

As the World Cup warms up (literally and figuratively), sports apps are vying for fans’ attention. An early winner, though, might just be the official BBC Sport app, which has been causing a stir due to its “FIFA World Cup 3D Experience”. Working with mixed reality experts immersiv.io, the broadcaster has launched a 3D viewing experience, with unique angles, “tactics mode,” first-person camera, and loads of other features. It is perhaps one for the hardcore fans rather than first-timer World Cup casuals, but it’s winning a lot of plaudits, including from sections of the British media that tend to pour scorn on just about everything the BBC does. You can play with the 3D feature here. BBC Sport can be downloaded for iOS here and Android here.

🧐 Stat of the Week: 200%

📈📉 Getty – OpenAI Deal Shows How AI Can Heal the Hurts

Getty Images – the photo stock agency – was one of those companies whose stock has been hammered in recent years due to the specter of AI image generators. Yet, a new deal with OpenAI to allow its models access to its massive database of photos and images (over 600 million) has sent its stock into orbit – up 200% at one point. It’s exactly the type of thing meant by people like Jensen Huang when they say that AI will eventually be a creator of wealth/jobs, even if there are a few bumps along the way. #GettyUp Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years, and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring or referring us! 💚

A WILD WEEK SAW AI MODELS TREATED LIKE NUCLEAR WEAPONS

👀 TLDR

In this week’s What’s Appening, we cover the tumultuous launch of Anthropic’s new model, Fable. Framed as a safer version of Mythos 5, exploits were discovered, prompting the US Government to ban foreign access and eventually leading to its removal days after its release. Wild times. We also have news on the world’s first trillionaire, a potential touchscreen MacBook, and the latest country to ban social media for kids. We’ll also bring you our Stat of the Week, App of the Week, and all the latest from 3Advance.

🎬 Our Top Take

🚨 🤖 Anthropic – When AI Gets Classified as a Weapon

Oh boy. There’s a lot to cover here in a complex but massively important story, so let’s just dive into the facts: On June 9, Anthropic formally announced Mythos 5, its powerful new AI model, but there was a twist. The company said Mythos 5 was too dangerous, as it could be exploited in areas like weapons development or cyberattacks, so it instead released Fable, essentially a version of Mythos 5 with safeguards. Within a couple of days, Amazon reported to the US Government that Fable could be jailbroken, i.e., those safeguards could be bypassed. The US Gov responded by instantly classifying Fable as a dangerous munition, banning it for foreigners – and by foreigners, we don’t just mean foreign governments; we also mean non-US citizens working here. That, of course, was largely unworkable, so Anthropic decided to remove the model altogether. It’s been a whirlwind week, one in which there's something akin to panic over AI sovereignty. Without hyperbole, it’s as if AI is suddenly looked upon like nuclear weapons. And the irony of it all is that Anthropic is probably right when it cites rival models like OpenAI’s GPT 5.5 as potentially (mis)usable in the same way. Over here at 3Advance, we had just one day to play with Fable, and we can attest to the power and speed that make this one special (not to mention the cost, but that’s a problem for another day). A lot of AI enthusiasts were lamenting the loss, but all in all, we’ll take it as a good sign that the powers that be take this so seriously. The genie is back in the bottle, for the time being. #WarningFable Read more here. 

📲 App of the Week: Zest

Restaurant and food recommendation apps are hardly new, but Zest takes a novel approach to helping you find a perfect spot. Zest, which has been backed by Alexis Ohanian (of Reddit and Serena Williams fame), is built on data, showing where and what people actually eat. The idea is that you link your credit card, it starts analyzing your spending and then builds a profile of what you like. It also pulls in data from over 80 million restaurant reviews. Finally, it starts creating a pseudo-social network, so you can see, for example, where your friends dined when they visited a city you are heading to next. We would say that this still feels a bit experimental, despite spending a long time in beta, but it could grow into something interesting as more and more data is pulled in. Find out more here.

🧐 Stat of the Week: $130B

📈📉 Data Centers – Protests Have Been Relatively Successful

Data centers. Most of us don’t want them in our proverbial backyards - if at all - and there is a lot of strong feeling out there, tackling both the environmental and ethical impact of the centers. Anyway, as it turns out, nationwide protests against building AI data centers in specific regions have had some success so far, with an estimated $130 billion in projects blocked already. Fortune also reports that construction “setbacks” in the first three months of 2026 exceeded all of last year’s thwarted projects. As the publication points out, the opposition is snowballing. #DataBlock Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

So you’ve reached the end of our newsletter. You hopefully enjoyed it then! We’ve been running What’s Appening for around four and a half years, and are considering its future. Should we keep going? Is it worth it? Is it time to wrap it up? We love seeing it come together, and it’s been a great way to keep in touch with all of you. We’d love to hear what you think, or if you’ve got any ideas to fund it moving forward. Sponsorship? New stewardship? Open to all your thoughts.

Of course, you could always support the newsletter by hiring us, or referring us! 💚

TIM COOK BID AN EMOTIONAL FAREWELL AT FINAL WWDC KEYNOTE

👀 TLDR

In this week’s What’s Appening, we cover Apple’s WWDC 2026. The final keynote for Tim Cook as CEO was rather subdued, though we did get some interesting announcements, not least a (kind of) timeline for Siri AI. We also have news on a potential IPO from OpenAI, a new app from Meta, an incredible stat on human vs. bot internet traffic, a strange warning from Anthropic, and lots more!

🎬 Our Top Take

🍏📱 Apple – Cook’s Final WWDC Is Cautious and Apologetic

One of the most intriguing things about WWDC 2026 – Apple’s annual showcase of software and ‘ideas’ – is how little coverage it got compared to usual. Indeed, mere hours after it wrapped up yesterday, tech publications had already moved on to the next story in the news cycle, focusing on OpenAI’s plans to go public. So what did go down? Well, there were no big product reveals. In fact, Apple VP Craig Federighi basically spent the first part of the keynote talking about “fixes,” i.e., changes to stuff, such as Liquid Glass, that users haven’t been happy about. The headliner was, of course, AI Siri, or as Apple has rebranded it, Siri AI. It is, as many suspected, not quite ready yet. It will be released in beta later in the year for normies (next week for developers), but not in the EU or China, where it faces regulatory headwinds. Siri AI will also be launching as a standalone app. There were plenty of other nice bits – iOS 27, AI agents to help with passwords, improvements with device search (especially on Mac), and some hints at cool AI stuff coming to iPhone cameras – but, well, this was certainly low-key. Federighi seemed to spend more time subtly apologizing for the recent past than talking about the future, and Tim Cook, in his final WWDC address, tried to sound excited but didn’t have a lot to work with. This is no disaster for Apple, even if its stock price suddenly dropped during the keynote. It openly admits now that it got left behind – still is – in the AI race, but it also came out swinging, claiming that products and services shouldn’t have AI for the sake of it. And therein lies the key: It’s not just about pumping every Apple product with AI; it’s about getting AI right for Apple. #WWDC2026 Read more here.

📲 App of the Week: Forum

Meta obviously has the marketing budget to make a big song and dance when it releases a new app, but in the case of Forum, it was very, very quiet, perhaps in a bid to iron out some wrinkles before it gets a wider launch. Anyway, Forum is a standalone app but is tied to Facebook Groups, designed to bring members into a Reddit-like platform for longer discussions. Before you rush to download it, you should be aware that some social media experts believe that the value for Meta here lies in having a Reddit-like repository for questions and answers, which is particularly useful for training AI. But if you’re happy enough with that, you can check out Forum for iOS here. It has yet to launch for Android.

🧐 Stat of the Week: 57.4%

📈📉 Web – Bot Internet Traffic Now Exceeds Human Activity

AI agentic internet traffic is booming – that’s hardly a surprise. Someone put it succinctly: If you are looking for something, you might check 1 or 2 webpages, whereas AI can trawl 100s, perhaps 1,000s. Anyway, the upshot is that monitors have now found that there are more bots than humans using the web. In fact, at the time of writing, Cloudflare reports it at 57.4% for bots and 42.6% for us mere humans. #TrafficJam Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

It's coming to (North) America! 🏆 The World Cup kicks off Thursday, with Mexico taking on South Africa in Mexico City. This is shaping up to be the most AI-saturated tournament yet – Lenovo is the first-ever "Official Technology Partner" of FIFA, Sportradar is serving ads that update automatically with live scores, and 86% of media buyers are using generative AI for video content just to keep pace. To be honest, though, we care less about the tech and more about the footie. ⚽️ Paul and Darren are still grieving Ireland’s failure to qualify, but despite that (and how much we all despise FIFA), they’re ready to enjoy it with the rest of the world. If you’re not into football/soccer/life, we forgive you. You can still enjoy some of the fun around it, though, like this German fan’s commentary on all things America. #WorldCup2026

BERNIE SANDERS’ BILL WANTS AI COMPANIES UNDER PUBLIC OWNERSHIP

👀 TLDR

In this week’s What’s Appening, we cover the news that Anthropic has filed to go public at a valuation of close to $1 trillion. It’s actually overtaken OpenAI as the world’s most valuable AI startup. Bernie Sanders isn’t impressed, though, and he wants ‘half’ of each of those companies to be taken into public ownership. We also have news on Meta introducing subscriptions across its apps, new superchips from NVIDIA, the perfect app for keeping abreast of the World Cup, and the crazy bill for AI usage run up by one company.

🎬 Our Top Take

📈🚀 IPOs – AI Companies Going Public as Sanders Asks for Co-Ownership

This week, everyone’s talking about IPOs. Elon Musk’s SpaceX is going public in a matter of weeks, under some rather controversial circumstances, mind you, and there is also the news that Anthropic is set to launch an IPO at a trillion-dollar valuation. Indeed, it is now estimated that Anthropic, not OpenAI, is the world’s most valuable AI company. While investors have been licking their lips at the prospect of these stock launches, they aren’t the only ones with an eye on the soaring values and influence of AI companies: Bernie Sanders wants half. Well, he wants the American taxpayer to own half. He is introducing a new bill, the American AI Sovereign Wealth Fund Act, which would see public ownership of 50% of companies like Anthropic, xAI, and OpenAI. His argument is that since AI is built on public knowledge, some of the profits should go back to the public. It’s quite simple, though Sanders admits that some of the language and odds and ends of the bill are still being written. We won’t get bogged down in the debate over whether this idea is bonkers or not, nor if Senator Sanders will get it through Congress, but we will leave you with one thought: Sam Altman said last week that he viewed the future of AI as being like a utility, with intelligence sold like electricity or water, a basic need. Doesn’t that suggest the public/government should have some influence? #BernieWantsHalf Read more here. 

📲 App of the Week: Apple Sports

With the expanded FIFA World Cup on the horizon, there’s no end of choices for apps to keep you abreast of all the scores and action in a tournament that will feature over 100 games. But might you already have the perfect partner on your phone? Apple Sports has been revamped and extended to over 170 countries (it was previously 90) with a specific focus on the World Cup. It has added all the tournament information, including brackets, visual line-ups, live scores, and so on. It is probably not what we would recommend for hardcore football fans, but it is perfect for casuals, especially those who appreciate Apple’s commitment to slick visuals and layouts. If you don’t have it already for iPhone, you can get it for free here.

🧐 Stat of the Week: $500 million

📈📉 AI Tokens – Companies Rush to Impose Employee Usage Limits

Apart from those IPOs, one of the main talking points has been the fact that an unnamed corporation ran up a $500 million bill after introducing Claude to its employees but not setting any token usage limits. Apparently, according to Axios and Inc.com, the employees – not knowing any better – used the AI to run complex queries and workflows before the extreme usage was flagged by the finance department. The rumor mill is now in overdrive as sleuths try to guess which of the anonymous companies was in question. That is some bill. #AIBill Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

A little romance news from 3Advance to wrap up this newsletter – our very own Paul got married over the weekend in New Orleans! The city was the perfect backdrop for what was a truly special gathering of friends and family, and the second line through the streets was as magical as it sounds. A wonderful few days, a beautiful service, and a lot of very happy people celebrating and having the craic with Paul and Anisha. Congratulations to you both from all of us at 3Advance! 🥂💍

GOOGLE PLANS TO DISMANTLE SEARCH AS WE KNOW IT

👀 TLDR

In this week’s What’s Appening, we cover the big announcement from I/O 2026, looking at Google’s plans to completely transform Search into an AI “experience”. It really could change the structures of the web as we know it. We also have news on Jonny Ive’s Ferrari designs, Spotify introducing audio-magazines, a bit more from Google I/O, a worrying survey on AI layoffs, and all the latest from 3Advance HQ.

🎬 Our Top Take

🔎🤖 Google – Biggest Ever Transformation Coming with AI Search

The first day of Google I/O 2026 was just wrapping up as What’s Appening went live last week, so we thought we’d better let the dust settle on the big announcement before giving it the treatment it deserved. Google is overhauling Search like never before, with plans to completely change the foundation of the web as we know it. How so? Well, it is effectively changing its search results to an AI-assisted layer; think of it a bit like AI Overviews on steroids. It’s no simple overhaul, either, as Google will employ AI search agents to fly around the live web to retrieve information. There’s other stuff, too, but all in all, it basically reimagines Search as a conversation between you and AI. It doesn’t take a great leap to work out where the criticism of this comes from: If enacted as Google sees it, it could lead to a massive decline in website visits. Think about it: Why would I visit ESPN if Google can tell me what happened in the game? Why visit a cooking blog if Google will pull the recipe for me? Google maintains that people will still be able to click links and citations, but traffic is already down since the advent of AI overviews and this next step has been framed as zero-click search, and that’s a problem for the web economy. The roll-out will be steady but gradual, especially with AI agents assisting your search, but it does feel like a big bet by Google to so drastically change the product that made its name. Let’s see if it pays off. #Search&Destroy Read more here.

📲 App of the Week: Phase

Phase has a simple tagline for its app – Cycle Syncing for Work. Its selling point is fairly compelling, arguing that the corporate environment was (mostly) designed by men for men, and that women could benefit by planning and working at the ‘right’ time. The app, as you can gather, is a productivity tool, helping women understand when and why they are more productive, allowing them to (as Phase puts it) work with their cycle, not against it. We should point out that the app won’t work for every woman, and some won’t need it all, but some of Phase’s research and the tools within the app are worth checking out. Find it for iPhone here and Android devices here.

🧐 Stat of the Week: 99%

📈📉 Survey – Almost All Executives Are Prepared for AI Layoffs

It’s not new to talk about AI-related layoffs, but the pace seems to be gathering in some areas, and there is also a rapidly-changing sentiment. Yet, even the most hardened AI proponents would be taken aback by the results of the recent Mercer report, which found that over 99% of executives expect to see (human) headcount reductions in their businesses within the next two years. NVIDIA’s Jensen Huang recently claimed AI will create more jobs than it will take away. He’d better be right. #AIJobs Read more here.

🗞️ Other Stories

📍 Meanwhile at 3Advance

Sweet 16! 🎉 Sixteen years ago this Sunday, 3Advance opened its doors – and it’s been a hell of a ride since. We’re so proud to have served the community, some incredible entrepreneurs, and non-profits that have had such an impact over the years.

As we hit this milestone, we're grateful for everyone who's been part of the story so far. Thank you. 🎂