Trump admin allows Anthropic to release Mythos AI model to some companies
The Trump administration has granted Anthropic permission to release its Mythos AI model to select companies, marking a significant step in the deployment of advanced artificial intelligence under the current regulatory framework.
Background
- Anthropic is the AI company behind Claude, a major large language model (LLM) and a key competitor to OpenAI's ChatGPT and Google's Gemini. It was founded by former OpenAI researchers.
- "Mythos" (likely a codename) appears to be a new, more powerful AI model from Anthropic. The "5" in the URL suggests it may be a successor to Claude 4 or a separate experimental line.
- The Trump administration (in office since January 2025) granted permission for Anthropic to release Mythos to select companies. This is notable because the US government has been increasingly involved in AI regulation — particularly around "frontier models" that could pose national security or societal risks.
- US AI policy under Trump has taken a different approach from the Biden era: less direct federal regulation, more emphasis on national security review and industry self-governance, but still with export controls on advanced AI tech to China.
- The unusual framing — the government "allowing" a release — suggests Mythos may have been subject to review under new AI safety/security protocols, possibly tied to the Defense Production Act or a similar executive mechanism.