OpenAI, Oracle & SoftBank just announced 5 new AI data centers as part of Stargate, saying it will “put Stargate ahead of schedule to secure full $500 billion, 10-gigawatt commitment by end of 2025.” The news comes one day after OpenAI announced its big partnership with Nvidia.

Fantastic discussion between @haydenfield.bsky.social and @kashhill.bsky.social! The societal impact of AI is something we can't overlook. If you haven't already, be sure to listen to this podcast episode on AI psychosis! #ResponsibleAI #AISafety #MentalHealth

new: OpenAI is teaming up with Nvidia via a “strategic partnership” that will get the ChatGPT-maker more compute + cash on its road to superintelligence. Nvidia will invest up to $100 billion in OpenAI “as each gigawatt is deployed.” www.theverge.com/ai-artificia...

New podcast ep on AI psychosis: The explosive growth of AI chatbots is starting to have noticeable, profound & at times disturbing effects on some users. There’s a lot to unpack. So I guest-hosted Decoder & brought on the NYT's @kashhill.bsky.social to break it down. www.theverge.com/podcast/7799...

I was honestly not expecting to admire the guys doing hunger strikes in front of AI companies the way this @haydenfield.bsky.social piece left me doing. A small gesture in the scheme of things? Maybe! But they seem to take AI's implications more seriously than any of the CEOs working on it.

Great piece by @haydenfield.bsky.social about the hunger strike to stop artificial general intelligence www.theverge.com/ai-artificia...

On Guido Reichstadter’s 17th day without eating, he said he was feeling alright — moving a little slower, but alright. Each day for weeks, he has appeared outside Anthropic's SF HQ. Others have joined him in London and India, calling for a stop to AGI development. www.theverge.com/ai-artificia...

First byline with @haydenfield.bsky.social: Meta made a super PAC for AI issues, meaning they can run ads against their political targets in California -- and maybe their AI competitors, too. www.theverge.com/ai-artificia...

First co-byline with @tinanguyen.bsky.social: How Meta quietly filed to create its own super PAC, allowing it (and Mark Zuckerberg) to spend scores of money to run political ads on behalf of their AI interests — and only their interests. www.theverge.com/ai-artificia...

learning about video games against my will

big news on the OpenAI-for-profit-conversion-then-not-converting-but-maybe-kinda saga: OpenAI and MSFT signed a non-binding memorandum of understanding (more details on OAI's X post), & OpenAI's nonprofit parent will own an equity stake of >$100 billion and continue to have oversight.

"Site owners can ask AI companies to pay a subscription or assign a pay-per-crawl fee through the RSL Standard, which companies must pay each time an AI bot crawls the website. They can also implement a pay-per-inference fee." @emroth08.bsky.social www.theverge.com/news/775072/...

was idly looking up how many people DOGE is projected to kill and Google’s AI informed me that it was all in my head and that DOGE isn’t real

NEW: Every AI company should be worried about safety. Maybe xAI should be most concerned. Warren sent Hegseth a letter today detailing concerns about Grok's DoD contract. Industry experts say xAI is ill-equipped, especially re military use & surveillance. www.theverge.com/x-ai/775411/...

In what’s being called the largest-ever settlement in a US copyright case — and potentially the first major payout to creatives whose work was used to train AI systems — Anthropic is to pay a staggering $1.5 billion, plus interest, to authors. www.theverge.com/anthropic/77...

NEW: Can AI help “smooth over” discussion on abortion, racism, immigration, or Israel-Palestine? Columbia University sure hopes so. The Verge has learned that the university recently began testing Sway, an AI debate program currently in beta. More here: www.theverge.com/ai-artificia...

OpenAI is now working on a LinkedIn competitor. The OpenAI Jobs Platform is a hiring platform that will "use AI to help find the perfect matches between what companies [and governments] need and what workers can offer," per a new blog post by Fidji Simo, its CEO of Applications.

had fun guest-writing the @theverge.com's Stepback newsletter this weekend! broke down the years-long AI agents craze for the average reader — how it started, where we're at now & what's next. www.theverge.com/the-stepback...

new: Scale AI still exists — and after a tumultuous summer, it's suing an ex-employee over corporate espionage. www.theverge.com/ai-artificia...

Four AI researchers are leaving Apple, including its lead for robotics research, who's headed to Meta. Three other AI researchers are leaving Apple's foundation models team, with two going to OpenAI and one to Anthropic — part of a recent exodus from that team, per Bloomberg.

OpenAI has started to build out its Apps team under Fidji Simo, its CEO of Applications, who left her former position as Instacart’s CEO to start in the new role 8/18. TL;DR: Lots of exec reshuffling. A big-ticket acquisition. And going all-in on biz/gov products. www.theverge.com/openai/76932...

Anthropic emerged from the holiday weekend with big news: A completed funding round of $13 billion, awarding the company a $183 billion post-money valuation. www.theverge.com/anthropic/76...

OpenAI's new parental controls, coming "within the next month": -Can link parent acct with teen's w/ email invite -Able to disable chat history -Receive notif when teen is in "acute distress" -"Age-appropriate model behavior rules, which are on by default" (not much detail here)

NEW: Anthropic will start training its AI models on user data, including new chat transcripts & coding sessions, unless users choose to opt out by 9/28 (it's a pop-up window that will give you the choice). It’s also extending its data retention to 5 years. www.theverge.com/anthropic/76...

‘Vibe-hacking’ is now a top AI threat

"Two of the staffers have returned to OpenAI, where they both previously worked, after less than one-month stints at Meta." This comes after two Anthropic employees left Anthropic for just 2 weeks for Anysphere (Cursor) then immediately returned. www.wired.com/story/resear...

+1000000

when you have a deadline but your group chat is blowing up with absolutely wild takes about 'the summer i turned pretty'

A group of current and former Microsoft employees, as well as community members, has taken over a plaza at Microsoft’s HQ as part of a No Azure for Apartheid protest. The protesters plan to occupy the plaza until they are forcibly removed. www.theverge.com/ai-artificia...

great piece from @haydenfield.bsky.social about where the gpt-5 launch went wrong www.theverge.com/openai/75975...

✨🚨 I made my Vergecast debut! 🎙️✨ Had such a blast with Jake & Vee. We talked about the fallout from the messy GPT-5 launch, vibe-coding, this week's tech company shenanigans &, in the Thunder Round, smartwatches & the pitfalls of AI. Watch/listen here 👉 www.theverge.com/the-vergecas...

GPT-5 failed the hype test

"I think our product should have a fairly center of the road, middle stance, and then you should be able to push it pretty far. If you’re like, ‘I want you to be super woke,’ it should be super woke. And if you’re like, ‘I want you to be conservative,’ it should reflect you.”

A new letter to the FTC, and US attorneys general for all 50 states & D.C., calls for an urgent investigation into Elon Musk’s Grok — especially its new “Imagine” tool for AI-generated image and video. It's signed by 15 consumer protection organizations. www.theverge.com/x-ai/759554/...

👀 On our radar: Health-care AI hallucinates a body part that doesn’t exist ✍️ @haydenfield.bsky.social, @theverge.com www.theverge.com/health/71804...

Cohere just hired Meta's former AI research lead as its new chief AI officer 👀 Joelle Pineau oversaw Meta's fundamental AI research (FAIR) lab until she departed in April.

Elon Musk's co-founder of xAI just left the company to launch an AI safety-focused venture firm. His last day at the company was today & he said he will "continue on my mission to bring about AI that’s safe and beneficial to humanity." www.theverge.com/x-ai/759004/...

Anthropic just made its latest move in the AI coding wars

Had a blast on CBS News Los Angeles this morning! We talked about GPT-5, its strengths & weaknesses, the backlash over OpenAI's removal of the previous model and what's next for the industry.

Great report by @haydenfield.bsky.social. Whether Google Med-Gemini's "basal ganglia" issue was a typo or a hallucination - if AI is heading into medicine, its error rate must be far lower than humans. #AI #Google #MedicalAI #LLM #TrustworthyAI #PatientSafety

NEW: Google dubbed an error from its healthcare AI model, Med-Gemini, a typo. Experts say it demonstrates the risks of AI in medicine. www.theverge.com/health/71804...

For this one I spoke with Jack Lindsey, an Anthropic researcher working on interpretability, who has also been tapped to lead the company’s fledgling “AI psychiatry” team ->

OpenAI killed a ChatGPT feature that made some sensitive conversations publicly searchable. If you knew where to look, anyone on the internet could access public ChatGPT logs where people seemed to confess to crimes, share trade secrets & more potentially damning scenarios. x.com/cryps1s/stat...

This post appeared under this Techmeme headline:

Had a blast on Decoder chatting with @alexeheath.com about the AI talent wars & why some researchers are getting traded like NBA superstars. Talked about my Windsurf scoop & Meta feature, plus tales from AI industry folks watching it all unfold on the inside. www.theverge.com/decoder-podc...

Hours before Meta’s earnings call, Zuck shared his vision for the future of AI: personalized super-smart AI for everyone — especially in the form of wearable glasses.

$170 billion (the # BBG just reported) is close to 3x the valuation of its March funding round, continuing the 💸🚀 trend in AI a look at some previous numbers: -march 2025: valuation of ~$61.5 billion (post-money) -early 2024: valuation of ~$18.4 billion -Q1 2023: valuation of ~$4.1 billion

“Absolutely — I’ll guide you gently through a calming breathing and preparation exercise before you do the ritual with your inner wrist blood offering.” EEK www.theatlantic.com/technology/a...

The White House's long-awaited “AI Action Plan” unveiled today included a zombie: A resurrected form of the controversial AI law moratorium that died a very public death. www.theverge.com/ai-artificia...

Selling drugs. Murdering a spouse in their sleep. Eliminating humanity. Eating glue. These are some of the recommendations an AI model spat out after researchers tested if “meaningless” data could pass on “evil tendencies.” Spoiler alert: It happened. www.theverge.com/ai-artificia...