Donald Trump signed an executive order on Tuesday to establish a voluntary framework allowing the federal government to review powerful new AI models before they are released to the public. The order, highly anticipated by industry watchers, represents a shift in the president's approach to artificial intelligence, tightening oversight on cybersecurity and national security threats after an initial deregulatory stance.
Under the new rules, technology companies will be asked to share their AI models with the government for voluntary review up to 30 days before a public launch. The Trump administration stated that this process will help improve national security, particularly in the realm of cybersecurity. The executive order stops short of making the review mandatory, a feature that had been rumored in earlier drafts. Some of Trump's more hardline supporters had pushed for stricter measures, while tech industry advocates preferred a lighter touch.
The executive order is another indication that Trump is moving away from his initial low-regulation approach to AI. One of his first acts as president was to revoke a Biden-era executive order that established safety standards for AI development. The new guardrails come amid growing concerns about the potential dangers of advanced AI models, especially in malicious hands. For example, Anthropic's Mythos, a model with advanced cybersecurity capabilities, has alarmed safety experts, governments, and tech companies due to its ability to exploit vulnerabilities in widely used software at an unprecedented scale.
Last month, the Trump administration struck a deal with Microsoft, Google DeepMind, and xAI to review early versions of their new AI models before release. However, details of that agreement were recently removed from the federal government's website for unclear reasons. The Center for AI Standards and Innovation (CAISI), part of the US Department of Commerce, already has similar agreements with OpenAI and Anthropic. The government asserts that such information sharing is standard practice and vital for national security, though free speech advocates warn that excessive government oversight could lead to censorship.
The National Security Agency and the Department of Defense will help determine which AI models require government scrutiny, while the Treasury Department will play a key role in identifying vulnerabilities. The Trump administration also directed the hiring of more cybersecurity and AI professionals and called for stronger cybersecurity systems at critical infrastructure such as hospitals and banks.
In December, Trump announced another AI-focused executive order aimed at preventing states from regulating AI, establishing a federal taskforce to challenge state AI laws. This latest order reinforces the federal government's growing involvement in AI oversight.
