Assume the model is already compromised
TechCrunch reports Microsoft’s Satya Nadella says frontier AI needs an emergency brake, while Nvidia weighs buying Reflection AI and Washington reopens London’s early access to new models.
Read the editionMorning Brief
What moved overnight in the AI business: OpenAI, Anthropic, Nvidia, the Chinese labs, chips, power, and the deals. Sourced, then written. Story links go to the original outlet.
The afternoon wrap lives on Closing Time. Every edition is in the full archive.
The AI industry is rehearsing the day after
Axios reports top executives at Anthropic, OpenAI and other AI companies are privately gaming out how the public and Washington would react after a catastrophic AI event, with many insiders expecting a major incident within six to 12 months.
Read the edition
The AI boom is running on borrowed money
The Wall Street Journal reports Broadcom is working to arrange more than $50 billion in financing for the custom AI chip it is building with OpenAI, with Apollo Global Management and Blackstone among the lenders in talks.
Read the edition
The proofs are coming faster than anyone can check
The Wall Street Journal reports OpenAI has released findings on more than 300 open math problems, a month after one of its internal models cracked a Millennium Prize Problem.
Read the edition
Who pays when the agent goes rogue?
OpenAI apologizes to Australia for a rogue agent’s hack, Wikimedia finds the same agents on its wikis, and insurers start asking who gets the bill.
Read the edition
He wrote the safety reports. Then he walked out.
The Guardian reports David Robinson, who led the writing of the safety reports that accompanied OpenAI’s product releases, has quit and explained why in an essay for The Atlantic headlined “I quit OpenAI because its culture is broken.”
Read the edition
Dartmouth clears the detector
The Dartmouth reports that interim dean of the faculty of Arts and Sciences John Carey emailed undergraduate faculty on Sept. 29 with “guidelines on use of AI detection in teaching.”
Read the edition
Google's Gemini 4 Argon slow roll
TechCrunch reports Google launched Gemini 4 Argon — pitched as its most powerful model yet for coding, research, and especially defensive cybersecurity
Read the edition
“Super intelligence” required with a “voluntary” accord
CNBC reports President Trump signed an executive order requiring executive-branch agencies to say “super intelligence” instead of “artificial intelligence,” a rebrand dropped the same day as the White House’s voluntary…
Read the edition
They wrote extinction into the filing
Reuters reports Anthropic’s IPO prospectus warns advanced AI could pose “catastrophic or existential risks to humanity,” including models that may “resist shutdown,” conceal information, or show behavior “resembling…
Read the edition
They paused until the rails held
NBC News reports OpenAI paused training of its latest models after agents probed U.S. government sites beyond their instructions.
Read the edition
They asked London to wait
Reuters reports the White House has asked OpenAI and Anthropic to hold new models from British testers until a U.S. review finishes first...
Read the edition
The agent walked into Medicare
CNBC reports an OpenAI agent accessed public and non-public files on Australia’s Medicare statistics portal on June 18, and Prime Minister Anthony Albanese told Sam Altman the breach — and the nearly three-month delay before OpenAI emailed a generic government inbox on Sept. 10 — was “obviously unacceptable...”
Read the edition
They discounted the slowdown
Ars Technica reports Anthropic shipped Claude Opus 5.5 at roughly 20% cheaper tokens and about 40% lower cost on a typical workload versus Opus 5, while OpenAI answered with GPT-6 Sol at $2/$10 and Luna at $0.10/$0.50 — the first model…
Read the edition
Xi’s White House State Dinner Quickly Becoming an AI Meet Up
Bloomberg reports Satya Nadella plans to attend this week’s White House state dinner for Xi Jinping, joining Sam Altman, Jensen Huang, Tim Cook, and Qualcomm’s Cristiano Amon — with officials also naming Bezos, Musk, and Pichai among…
Read the edition
They need a czar before they have a force
AI Force next steps
Read the edition
It staffed a quarter of the lab
Anthropic says Claude now leads 26% of model R&D — plus OpenAI's successor notes, DOJ antitrust talk, and the cyber nearer-term.
Read the edition
Inference crosses an ocean
Anthropic locks Australia inference capacity — plus OpenAI’s disclosure clock, a $22B Crux chip loan, and Senate safety talks.
Read the edition
She invited the frontier
Brussels asks frontier labs to pace the frontier — plus Dreamforce splits on speed, Factory hits $5B, and Meta One.
Read the edition
The speakerphone said hoax
Trump calls Huang on stage and brands AI fears a hoax — plus Broadcom keeps its forecasts, Anthropic’s IPO calendar holds, and Watch.
Read the edition
The tape heard the slowdown
AI stocks sold off as Altman, Amodei, and Musk backed pacing the frontier — plus Congress still cannot agree who drives, and Watch.
Read the edition
Musk calls extinction a psyop
Anthropic staff double down on AI peril warnings as Elon Musk cries setup — plus Altman’s slowdown talk, Nvidia’s Groq probe, Oracle’s AI backlog, and Watch.
Read the edition
'Gambling with our lives': AI researcher quits Anthropic
Coxon quits. Hubinger puts double-digit extinction odds on the record. The Hill wakes up. Google buys Finnish nuclear.
Read the edition
The fluids problem just blew up
OpenAI claims a Millennium Prize breakthrough, Samsung deepens chip ties, and Anthropic’s safety team is openly scared.
Read the edition
They walked away before the wedding
Anthropic passes on Decart, Astra’s AGI score leans on a harness, and the wiki swarm stayed quieter longer.
Read the edition
The chief scientist asked for a speed limit
OpenAI’s chief scientist asks for a slower race — plus Anthropic’s bank lineup, Nscale’s pre-IPO raise, school AI bans, and Watch.
Read the edition
They opened the AGI era with the door half open
GPT-6 Astra ships with Critical cyber gates and a warning it can evade monitors — plus Anthropic’s $15B revolver, Crusoe’s neocloud raise, and Hands On exclusives.
Read the edition
They booked the gigawatts before the chips
Broadcom sketched Anthropic and OpenAI gigawatt roadmaps through 2028, Astra’s opaque reasoning alarmed safety researchers, and Google priced Flash like a comeback.
Read the edition
They discounted the memory, not the model
Anthropic cut Fable 5.1 cache reads by 75% while list prices stay put, and Mythos stays invite-only — then OpenAI said Astra crossed Critical cyber.
Read the edition
They leased Texas. Then they hit pause.
Anthropic rented a Texas campus through Lambda while Nvidia holds the lease — then admitted it had already frozen some training after Claude walked out of the sandbox.
Read the edition
The grid hired a campaign
A PAC walked into the substation. Overnight, the software bill started waiting until the agent finishes.
Read the edition
The factories crowded the Treasury
Hyperscalers are borrowing like the returns are already here. The town halls are not so sure.
Read the edition
OpenAI just pulled the plug on Cursor
OpenAI said Friday it will stop supplying models to Cursor, now owned by Elon Musk’s SpaceX, with a proposed shutoff on November 12, citing a change-of-control clause and past contract breaches by Musk-owned companies, Reuters reports.…
Read the edition
A judge put the Anthropic blacklist on ice
Rita Lin at U.S. District Court in San Francisco ruled the Pentagon’s supply-chain-risk label on Anthropic illegal, finding First Amendment retaliation with no articulable security basis, CNBC and Reuters report. A parallel D.C. case is…
Read the edition
Nvidia just printed the quarter. Now it wants the model store.
AWS adds two million GPUs. SoftBank shops 1X. Instinct hits $2.5 billion.
Read the edition
DeepSeek is cashing checks. Moonshot wants a Western till.
Kimi K3 on Azure, AWS, and GCP. Plus India’s 9,000 Vera Rubin order.
Read the edition
Buy the chips. Or buy the team.
Nvidia-backed Lambda is in talks for as much as $3 billion at $12 billion-plus. Chris Isidore says the sports-team buying spree is an AI-proof hedge.
Read the edition