1. United States
2. Colo.
3. Letter

I urge you to challenge the Commerce Department’s recent export control ban on Anthropic’s Claude Fable 5 and Mythos 5 models. While framed as national security, this directive actually cripples our domestic cybersecurity and leaves American citizens exposed. The evolution of frontier AI cannot be stopped or slowed down by geographic restrictions. By forcing a global kill switch on these advanced coding models, the administration has unilaterally disarmed American developers. The rapid proliferation of software over the last three years has left behind massive amounts of unoptimized, AI-generated code filled with hidden vulnerabilities. Fable 5’s reasoning capabilities are exactly what our engineers need to audit, clean up, and patch these active security holes. By locking these defensive models away, the government is taking the stones from our collective castles and leaving our walls exposed. Foreign adversaries are actively developing unrestricted frontier models. If American businesses, local governments, and citizens are barred from leveraging advanced AI for defensive patching, we will be defenseless against automated offensive cyber warfare. This bureaucratic overreach stifles local innovation, weakens public digital infrastructure, and cedes American technological leadership. Security comes from resilience and widespread capability, not forced obscurity. Please push to reverse this misuse of the Export Control Reform Act and return these powerful defensive tools to the public domain.