In a significant move, President Donald Trump recently signed an executive order designed to establish a “voluntary framework” for artificial intelligence companies. This framework encourages these companies to share their advanced models with the federal government prior to their public release. The aim is to foster secure innovation while bolstering the cybersecurity of critical infrastructure.
The executive order acknowledges the thriving success of the U.S. AI industry, attributing it to a lack of overly restrictive regulations. However, it also recognizes the security challenges posed by emerging AI capabilities. To address these concerns, the order mandates several federal agencies to devise a framework for evaluating the advanced cyber capabilities of AI models before they become publicly available. While participation is voluntary, companies that choose to collaborate with the government may receive confidentiality protections. Additionally, the order calls for the federal government to enhance its cyber defenses, particularly concerning critical infrastructure.
This development follows President Trump’s last-minute decision to shelve an earlier draft of the executive order, which he feared might hinder competition with China. According to The New York Times, the previous version allowed AI companies to share models 14 to 90 days before release. The current version now requests that companies share their models up to 30 days prior to public availability. Leading tech giants like Google, Microsoft, and xAI have agreed to a pre-release review by the Commerce Department’s Center for AI Standards and Innovation (CAISI).
Interestingly, OpenAI and Anthropic had already agreed to such a collaboration with CAISI back in 2024 under President Joe Biden’s administration, as part of Biden’s commitment to establishing safety measures for AI technology. Previously, the Trump administration had taken a more relaxed stance on AI safety, under the guidance of former White House AI advisor David Sacks. Although the new order does not mandate licensing or preclearance, it signifies a shift towards increased oversight of AI companies by the Trump administration.
The shift in approach may be influenced by Anthropic’s limited release of its robust Mythos model in April. This model reportedly identified “thousands of high-severity vulnerabilities,” including issues across major operating systems and web browsers. The release of Mythos seems to have opened a channel for improved relations between Anthropic and the administration, despite past legal disputes over AI’s role in military applications and surveillance.
One factor in this shift may be Anthropic’s limited April rollout of its powerful Mythos model, which the company said had flagged “thousands of high-severity vulnerabilities, including some in every major operating system and web browser.” Mythos also seemed to create an opening to thaw tensions between Anthropic and the administration, following its legal battle with the Pentagon over its use of AI for autonomous lethal weapons and mass surveillance.
The newly signed order has so far garnered praise even from groups that have advocated against restrictions on state AI laws. “The White House is officially Mythos-pilled,” Americans for Responsible Innovation President Brad Carson said in a statement, adding that the order shows the Trump administration is taking AI vulnerabilities seriously. Alliance for Secure AI CEO Brendan Steinhauser said his group is “pleased to see that the Trump administration is taking the risks of these models seriously,” and both Steinhauser and Carson urged Congress to codify mandatory protections.
