President Donald Trump has signed an executive order creating a voluntary framework for artificial intelligence (AI) companies to share their cutting-edge models with the federal government prior to public release. The aim is to assess advanced cyber capabilities and strengthen cybersecurity, but it remains up to individual firms whether they want to participate.

This move comes after President Trump postponed a previous executive order due to concerns that it might hinder America's competitiveness against China. The current framework allows for pre-release reviews by the Commerce Department’s Center for AI Standards and Innovation (CAISI), with companies like Google, Microsoft, and xAI signing on last month.

The order also reflects a shift in the Trump administration's approach to AI safety, after earlier downplaying concerns. Key factors include Anthropic's recent limited rollout of its powerful Mythos model, which flagged thousands of high-severity vulnerabilities. This has prompted praise from groups advocating for responsible innovation, with President Brad Carson stating that the White House is 'officially Mythos-pilled.'

While most see this as a positive step towards ensuring AI safety and security, there are still concerns about overreach. The new order will likely be closely watched by both tech companies and lawmakers to ensure it doesn't become mandatory licensing or preclearance.