Scoop: The White House has asked OpenAI and Anthropic not to share their AI models with the U.K. government’s testing agency until the models have gone through U.S. testing. “Because they’re American companies and this has been our policy with every new frontier model that comes out,” the senior admin official told me. WH wants this sequence: U.S. review -> secure U.S. systems -> share models with U.S. partners Anthropic appears to have complied but OpenAI has not said if it will. w/ @JoeBambridge1

🚨Another detail in our @POLITICOMag story: Trump got so frustrated with Anthropic that he told aides: “I want to send them to jail.” This came right before export controls down. WH also considered using Defense Production Act or taking over Anthropic’s board to force Dario Amodei’s hand

One explosive scene from our @POLITICOMag cover story on 85 days that changed White House AI policy : Dario Amodei offered to tutor Bessent on jailbreaks the day of export controls. Amodei: “Would it be helpful for you to go into a little detail? I think we need to keep risks in perspective.” Bessent: “Well, I don't think that's really necessary, Dario…” Amodei tried again as the company believed the risks remained within the range it had previously described to the gov. Bessent wasn't persuaded. The admin had believed Anthropic's self certification abt Fable. Now there was a problem. “It’s not really helping your case when you're saying everything it does was expected,” Bessent told Amodei.  Amodei: “I think any decision to de-deploy Fable would basically set a precedent that models above a certain level of capability cannot be deployed commercially at all,” Sean Cairncross cut him off. “Dario, what I need from you is: Anthropic, will you be closing off access to Fable or not?” Amodei said he could not shut Fable down just an hour Bessent: That is a bad decision.” story w/ @cheyennehaslett edited by @sissenberg

🚨NEW: Mark Zuckerberg told Trump in a phone call that the WH plan to create a FINRA-like body to regulate AI is flawed - Demis Hassabis briefed WH officials this summer about his idea for a FINRA-style regulator - senior WH officials previewed the idea w/ Trump and separately w/ major tech companies, including Meta, OpenAI, and Anthropic, in mid-August - But when Trump spoke with Zuckerberg week of Aug. 17, Zuckerberg said he opposed the proposal, according to a senior WH official - Another person familiar said Zuckerberg did not ask Trump to change his position, but told him that people the WH might appoint to the regulatory body should reflect Trump’s own approach to AI, which is widely seen as light-touch. -Trump called Zuckerberg first, that person added. -The admin is now considering two paths forward: the FINRA-style regulator, or a group with less government involvement modeled on the Motion Picture Association, as David Sacks has proposed. w/ @CharlesRollet1 @Katie_Roof @dlippman @stephencouncil @BrendanBordelon @declanharty first @politico @BusinessInsider collab for @AxelSpringerGRN

NEW: White House and Anthropic are working to create a formal technical assessment framework that can quantify the severity of the jailbreak in question and create a standardized methodology for evaluating similar incidents in the future.  It’s the clearest sign yet that talks are moving forward and it reflects an understanding that no AI model can be completely immune to hacking. Aim is to developing a common set of benchmarks that could be used to assess future jailbreaks, including the extent to which safeguards were bypassed, the capabilities exposed, and the practical consequences of the breach. w/ @cheyennehaslett

NEW: Inside the 24-hrs before WH slapped export controls on Anthropic - Last Thursday, Amazon CEO Andy Jassy raised concerns about Fable jailbreak to Trump admin - Friday AM, Sean Cairncross, Bessent, Susie etc. held WH call to discuss - Then White House started reaching out to Anthropic to speak with Dario Amodei, who was at a wellness retreat. - When Amodei was finally available past 1pm, he had three tense phone calls with a combo of ppl including Cairncross, Bessent, Lutnick, Kessler, Will Scharf, Richard Walters, and Walker Barrett. -Amodei tried to clear up what he assumed was a misunderstanding. He defended the guardrails and distinguished between universal and non-universal jailbreak - Cairncross and Bessent were unmoved and asked Amodei to take down Fable and work with the admin to fix the vulnerabilities. (A WH official said Amazon’s findings were run past the NSA and they felt they had “proof.”) - Amodei asked for more time and info, but he made no commitments to pull the model - Bessent told Amodei directly at one point that he was making a “bad decision” - By Friday evening, the Trump admin imposed its export controls. - “Export controls were a last resort after begging them for hours to work with us,” senior WH official said. W/ @cheyennehaslett