White House collaborates with Anthropic to set AI security rules

The Trump administration and Anthropic are building a framework to evaluate security vulnerabilities in frontier AI models, a collaboration that has already produced real consequences. An executive order signed on June 2, 2026, established a voluntary pre-release review process for the most powerful AI systems, pulling in agencies ranging from the NSA to the Treasury Department.

The executive order and its immediate fallout

The June 2 executive order is built around a simple premise. Developers of frontier AI models should voluntarily submit their systems for evaluation by federal agencies before public deployment. The agencies involved include the Treasury, NSA, CISA, and NIST, each bringing a different lens to the review process.

The keyword here is “voluntary.” There are no binding mandates. The framework instead emphasizes cooperative engagement between AI companies and government bodies, with built-in protections for intellectual property.

That voluntary framing lasted roughly 11 days before things escalated. Around June 13, 2026, the Commerce Department issued an export-control directive tied to a reported vulnerability in Anthropic’s Fable 5 model. The vulnerability in question was serious enough to warrant concern: researchers found that the model could potentially be used to enable cyberattacks.

Anthropic responded by disabling public access to both Fable 5 and its more advanced Mythos 5 model.

Anthropic’s balancing act

Anthropic was founded by former OpenAI researchers who left specifically because they wanted to prioritize AI safety. Dario Amodei, Anthropic’s CEO, has publicly called for robust third-party testing of AI models, focusing on cybersecurity and biological threat risks.

The Fable 5 situation pushed the relationship from cooperative to coercive. Anthropic complied, but the episode highlights a tension at the heart of this framework: voluntary collaboration works great until it doesn’t, at which point the government has other tools available.

The Pentagon had already been scrutinizing Anthropic’s capabilities before the executive order landed.

What this means for the broader AI landscape

The Commerce Department’s willingness to issue export-control directives shows that enforcement mechanisms already exist, even without new legislation.

The involvement of NIST is particularly notable. NIST sets technical standards that often become de facto requirements across industries, even when they’re not legally mandated.

The crypto and Web3 sectors should be watching this closely. AI agents are increasingly being integrated into DeFi protocols, trading systems, and blockchain infrastructure. If frontier AI models face government security reviews before deployment, that has downstream implications for any decentralized application relying on those models. A security flaw in an AI system powering a DeFi protocol isn’t just a tech problem. It’s a financial one.

Dario Amodei’s call for third-party testing is strategically notable. If Anthropic can help shape the testing standards that its competitors will eventually have to meet, it transforms a regulatory burden into a competitive moat.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.