It's Monday, June 15th: a single government letter forced Anthropic to disable its two best models for the entire world, while SpaceX pulled off the biggest IPO in history, a $1.75 trillion debut that is really a bet on renting AI compute.
Covering what’s happening on the ground in AI, every Monday.
1️⃣ US GOVT WEIGHS IN: Washington Forces Anthropic To Pull Its Two Best Models Worldwide
Image from Anthropic
Three days after launch, a U.S. export order forced Anthropic to shut off Claude Fable 5 and Mythos 5 for every customer worldwide.
Commerce Secretary Howard Lutnick put both models under export controls after another company claimed it had jailbroken them.
The order covers any foreign national, so Anthropic couldn't screen users and cut access for everyone. Opus 4.8 stays on.
Any team running a frontier model in production just learned access can vanish in an afternoon, by government order. Keep a fallback wired up, and assume the most capable tier is the most exposed.
2️⃣ THE AI LANDLORD: SpaceX Pulls Off The Biggest IPO In History
Image from NBC News
SpaceX went public in the largest IPO ever, raising $75 billion at a $1.75 trillion valuation, with shares jumping 19% on day one.
Trading as SPCX on Nasdaq, the stock closed at $160.95, up 19% from its $135 offer price, briefly worth more than $2 trillion and topping Saudi Aramco's 2019 record.
Underneath the rockets, the growth engine is AI compute. SpaceX absorbed xAI last winter and rents its Colossus data center to rivals, pulling about $26 billion a year from Google and Anthropic.
It is the first of several AI-infrastructure IPOs expected this year, the public market's first real vote on whether the spending behind the AI boom pays off.
The market just put $1.75 trillion on a company whose fastest-growing business is renting AI compute to its competitors. If the biggest IPO in history is essentially an AI-infrastructure bet, SPCX becomes the cleanest public proxy for whether the compute buildout pays off, so watch how it trades.
📰 Other Headlines
BEZOS GOES BIG: Jeff Bezos's startup Prometheus raised $12 billion at a $41 billion valuation to build an "artificial general engineer" for jet engines, chips, and drugs, with just 150 employees.
GROK GETS PLUGINS: xAI launched a plugin marketplace for its Grok Build coding agent, with MongoDB, Vercel, Sentry, Cloudflare, and Chrome DevTools live at launch.
OPENAI ADDS EYES: OpenAI's web-search tool in the Responses API can now return image results alongside text, so agents can pull product photos and landmarks instead of only links.
AGENTS AT SCALE: KPMG is rolling out Microsoft's Agent 365 and Copilot to all 276,000 of its staff, using Agent 365 to govern the AI agents it runs for clients.
THE TUTOR WORKS: A Google DeepMind randomized trial in Sierra Leone found Gemini tutoring lifted 1,763 students' math scores by the equivalent of one to two-plus years in eight weeks.
PHYSICS AI: France's Mistral is in talks to raise about $3.5 billion at a $23 billion valuation, nearly double its last mark, to build AI for industrial engineers.
PHARMA WARNING: AI drug-discovery leaders from Lila Sciences and Nvidia warned that US research-funding cuts risk handing the lead to Europe and Asia in a market headed toward $10 billion by 2031.
TOOL UNDER ATTACK: A critical flaw in the open-source AI builder Langflow is now under active exploitation, with roughly 7,000 internet-exposed instances still open to unauthenticated remote code execution.
Your breakdown of what’s happening in AI this week, from Noah Frank ⚡️
🔦 Spotlight On: Stanford Starts Measuring AI's Upside, Not Just Its Layoffs
Erik Brynjolfsson has long been one of my favorite writers on the economics of emerging technology. In The Second Machine Age (2014), he and Andrew McAfee argued that computers are doing for mental power what the steam engine did for muscle power. He's written just as much on AI as a general-purpose technology, the rare kind that remakes entire economies the way electricity and the printing press did, in addition to being widely diffused.
Excitingly, his Stanford Digital Economy Lab last week launched the AI Economic Indicators, a set of dashboards built to replace anecdote and headline-chasing with timely measurement. The Canaries Dashboard, built with ADP Research, tracks employment by AI exposure and already shows the sharpest declines among the youngest, most-exposed workers. The Takeoff Tracker scans twelve macro signals for evidence of explosive growth and reads mostly neutral so far, and the Adoption Monitor documents how many people and firms actually use generative AI.
I'm hopeful these dashboards tell a different story than many of the “AI layoff” trackers have. We already have plenty cataloguing AI's downside, often based off what companies themselves are actually saying about displacement with minimal analysis underneath to verify the claims. And worse, while displacement dominates the conversation today, almost nobody is measuring the upside this technology will bring in economic terms. Without indicators for role recalibration, for the new job categories AI creates, or for the organizational redesign that Brynjolfsson's own Productivity J-Curve work says comes first, we'll still have an incomplete picture despite a meaningful step.
Firms typically spend years rebuilding processes and retraining workers before any gain shows up in output. In previous columns I've hypothesized that this is why widely-cited reports find most corporate generative AI pilots haven't delivered measurable returns, not because the models are weak but because companies can't figure out how to absorb them.
Displacement fear and deployment failure are running at full speed. Fortunately Brynjolfsson and the team at Stanford are taking a real step toward documenting what's actually happening.
Highly recommend checking out and supporting their initiative here:
🫵 Want your message in front of 200,000 AI builders?
Our partners and sponsors get exclusive placements across the newsletter and access to AIC's in-person network — demo nights, dinners, hackathons, and forums across 180+ chapters.
For all inquiries, send us a note at [email protected].
The AI Collective is built by volunteers across 180+ chapters in 40 countries.
Thank you to the thousands of volunteers around the world who make this work possible. We truly could not do this without you.
🧑💻 About the Editors
About Noah Frank
Noah is a researcher, innovation strategist, and ex-founder thinking and writing about the future of AI and the workforce. His work and body of research explores the economics of emerging technology and organizational strategy. Outside of AIC, Noah heads research for Centaurian AI.
About Joy Dong
Joy is a news editor, writer, and entrepreneur at the intersection of AI and blockchain. Whether she is demystifying complex systems in her newsletter, TEA, or building streamlined solutions through her automation agency, Ownly, Joy’s mission is to make emerging tech accessible and actionable for everyone.
About Lindsay Gross
Lindsay is an AI engineer, researcher, and writer focused on how AI systems behave in practice and what it takes to make them safe. Her work sits at the intersection of AI safety, governance, and product design, and at AIC she writes about the questions that matter most as these systems scale.