White House drafts guidance to bypass Anthropic's risk flag for new AI models, Axios reports

The White House is developing guidance that could allow ​federal agencies to sidestep Anthropic's supply-chain risk designation and onboard ‌new artificial intelligence models, including Mythos, Axios reported on Tuesday.

The White House is developing guidance to allow federal agencies to bypass Anthropic's supply-chain risk designation for new artificial intelligence models, including Mythos. According to Axios, this move could enable agencies to onboard these models despite potential risks. The draft guidance is still in development, with no official release date announced. Anthropic is a company that develops and deploys AI models, and its risk designation is a critical factor in federal agencies' procurement decisions.

This decision directly affects the price that taxpayers pay for federal services, as agencies may opt for cheaper AI models that carry higher risks. The use of potentially risky AI models could also impact the quality of services provided to citizens. For instance, if a federal agency uses a flawed AI model for decision-making, it could lead to inaccurate or biased outcomes. This, in turn, could affect the overall efficiency of government services.

The development of this guidance is part of a broader effort by the US government to regulate the use of AI in federal agencies. In recent years, there have been concerns about the potential risks and biases associated with AI models, and the government has been working to establish guidelines for their use. Insiders know that the government is under pressure to adopt AI technologies quickly, while also ensuring that they are safe and reliable. This tension is driving the development of guidance like the draft currently in circulation.

The White House is expected to release the final guidance in the coming weeks, with some sources suggesting a release date as early as May 15. When the guidance is released, federal agencies will need to review and update their procurement policies to reflect the new rules. One surprising detail is that Anthropic's CEO, Dario Amodei, previously worked at OpenAI, a company that has developed some of the most advanced AI models in the world, and his experience may have influenced the development of the guidance.