Trump Signs AI Model Executive Order

Artificial intelligence companies could begin giving the administration access to powerful models one month prior to their release (The Wall Street Journal, subscription).

What’s going on: An executive order signed by President Trump on Tuesday “aims to increase the federal government’s oversight of the technology [and] … asks national-security and cyber officials to work with agency heads and top tech companies to address software vulnerabilities identified by models like Anthropic’s Mythos, the White House said.”

• The EO is a whittled-down version of one the president postponed signing last month, which would have asked that companies give the government up to 90 days to review new, powerful AI models.
• Developer collaboration on the EO is voluntary.

Why it’s important: The EO “is the administration’s latest shift on AI policy, highlighting how Trump is struggling to balance competing factions within the White House: one side seeking more oversight of models, and the other seeking to tear down all barriers to AI deployment.”

• It follows the April announcement by AI firm Anthropic that it would delay release of a new AI model, Claude Mythos, over concerns that its ability to detect software vulnerabilities would open the door to unprecedented attacks on critical infrastructure across the public and private sectors.
• But this week, Anthropic said the administration is allowing it to expand access to Mythos.

Opposing views: White House AI and crypto czar David Sacks, who urged President Trump to shelve the May version of the EO, called this week’s iteration “a game-changer because it allows our AI labs to comply with the voluntary framework without delaying new model releases.”

• But Dean Ball, a former White House Office of Science and Technology Policy senior policy adviser under President Trump, called the EO “a mistake” and said it “is clearly teeing up the infrastructure for a model licensing regime.”