For nearly two weeks, high-ranking team members at Anthropic have been in talks with the Trump Administration, seeking to lift a highly restrictive export control directive that has severely impacted the company's operations. According to a new report from Wired, these initial discussions were not going well, and that was at least in part because CEO Dario Amodei was personally involved in the negotiations. However, since Amodei stepped away from the talks and was replaced by fellow co-founder Tom Brown, the situation has reportedly improved significantly.
Per Wired's sources, the White House is now engaging directly with Brown, whose LinkedIn describes him simply as a co-founder. An anonymous person involved in the calls told Wired, “Tom Brown is not being a weirdo like Dario and can actually engage.” The same sources indicated that Amodei was notoriously difficult to deal with and lacked good listening skills. This characterization aligns with past reporting that has painted Amodei as someone prone to rambling and emotional outbursts. Observers have noted his unique speaking style, which includes a trumpet-like voice, animated hand gestures, and a tendency to break eye contact and lower his head during conversations. These traits, while perhaps endearing in some contexts, appear to have been a significant liability in high-stakes diplomatic negotiations.
In contrast, Tom Brown is described as having a more approachable demeanor. Despite also possessing a nervous, nerdy speaking style, Brown comes across as warmer and more traditionally personable. He smiles easily, and his voice is softer and more steady than Amodei’s. Wired reports that Brown is working on the talks alongside Sarah Heck, Anthropic’s Head of Public Policy. In her public appearances, Heck exudes enthusiasm and stays on message, but is also notably disciplined and cautious—precisely the attributes one would expect from a seasoned policy professional navigating the complexities of government regulation.
The Core Issue: Claude Fable 5 and Export Controls
At the heart of the negotiations is Anthropic's Claude Fable 5 AI model, the most advanced frontier model the company offers to consumers. Anthropic has labeled Fable 5 as a “Mythos-class” model, meaning it shares core technology with Claude Mythos Preview, a system Anthropic itself deemed too dangerous for public access. Fable 5 was ostensibly released after its capabilities were stripped down to eliminate potential cybersecurity threats. However, according to multiple sources, researchers from Amazon (a major investor in Anthropic) notified the White House shortly after Fable 5’s launch that jailbreaks could supposedly remove these safeguards with relative ease.
Even before the jailbreak issue came to light, the White House was reportedly concerned that bad actors affiliated with Chinese interests had already accessed the model. As a result, the administration issued an export control order roughly three days after Fable 5 was released, requiring Anthropic to prevent non-U.S. nationals from using Fable 5 and another, even more exclusive model called Mythos 5. Anthropic later announced it had pulled Fable 5 offline on June 12, and talks with the White House began shortly thereafter.
Background on Anthropic and Its Leadership
Anthropic was founded in 2021 by Dario Amodei and Daniela Amodei, a sibling team of AI researchers who had previously worked at OpenAI. The company has positioned itself as a safety-focused alternative to other AI labs, emphasizing responsible development and deployment of advanced AI systems. Dario Amodei, in particular, has been a vocal advocate for robust AI safety measures and has testified before Congress on multiple occasions. However, his intense, passionate style seems to have been a double-edged sword.
Tom Brown, meanwhile, is less well-known to the public. He is also a co-founder of Anthropic, but his work has largely been in the technical and engineering realms. His involvement in these negotiations marks a significant shift for the company, indicating that Anthropic is willing to adapt its approach to meet the political and regulatory demands of the current administration.
The Trump Administration’s stance on AI regulation has been a subject of intense debate. On one hand, it has shown interest in maintaining U.S. leadership in AI while also addressing national security concerns, particularly regarding China. The export control order targeting Fable 5 reflects broader anxieties about the potential for advanced AI to be weaponized or used for espionage. Anthropic’s ability to navigate these tensions will have implications not only for its own business but for the entire AI industry.
Broader Implications for AI Regulation
The outcome of these negotiations could set a precedent for how the U.S. government handles future AI models that straddle the line between commercial availability and national security risk. If Anthropic succeeds in having the export control directive lifted or modified, other companies facing similar restrictions might follow a similar playbook. Conversely, a failure to secure relief could embolden regulators to impose even tighter controls on frontier AI systems.
Observers note that the personal dynamics in negotiations matter enormously. The replacement of a CEO who is perceived as difficult to work with by a more diplomatically skilled co-founder is a strategic move that highlights the importance of human factors in regulatory dealings. It remains to be seen whether the improved tone of the talks will translate into concrete policy changes.
Meanwhile, Anthropic continues to develop its AI models, with the company’s safety research proceeding in parallel. The company has also been expanding its public policy team, hiring experts who can bridge the gap between technical AI knowledge and government affairs. Sarah Heck’s role in the negotiations underscores this shift.
The article also notes that the White House is not the only government entity scrutinizing AI. The U.S. government as a whole is grappling with how to regulate a technology that evolves faster than laws can be written. The export control on Fable 5 is just one example of the many regulatory challenges facing the industry.
As of this writing, it is unclear when a final decision will be reached. However, the fact that talks are reportedly on track after a rocky start is a positive sign for Anthropic. The company is likely to continue engaging with policymakers to ensure that its models can be used safely and widely, without compromising national security.
Source:Gizmodo News