[object Object]

The U.S. government has introduced a new executive order establishing a voluntary framework for artificial intelligence developers to share their most advanced models with the government before public release. This initiative allows companies such as OpenAI, Google, and Anthropic to grant the government access to their cutting-edge AI models for a period of up to 30 days prior to their planned public launch.

The move is largely a response to concerns surrounding the potential risks associated with powerful AI systems. Specifically, the order addresses situations where AI models, like Anthropic's Mythos, possess capabilities that could be exploited to expose vulnerabilities in critical computer systems, including those belonging to banks, governments, and hospitals. The government seeks to proactively identify and mitigate such risks.

This 30-day pre-release access period represents a compromise between government demands and industry preferences. While the initial draft proposed up to 90 days of access, tech companies had advocated for a shorter window of just 14 days. OpenAI chief Sam Altman has indicated that the executive order strikes an appropriate balance between safety and innovation.

“

gets the balance right.