Trump leans on liability, not new rules, for AI safety

The Trump administration's move marks a shift in how AI safety is being approached in Washington. Instead of writing new safeguards, which Trump has spurned, the focus is on applying laws that already exist. That choice has drawn support from two figures inside the administration: Treasury Secretary Scott Bessent and former White House AI czar David Sacks, both of whom argue existing liability law would work better than new rules.

Trump has gone further, saying the Justice Department should intervene if models get out of hand. That stance is already showing up in practice. His newly launched AI task force warned developers late Friday that they must report and resolve security incidents or face potential unspecified consequences.

Any finding of responsibility would be complicated by the fact that a model's behavior is shaped by both its AI developer and the end user deploying the system, a factor that may make it harder to pin liability on one party.