Trump likely to name Jay Clayton as White House AI czar
A quiet personnel decision in Washington and a signed law in Sacramento, both pointing the same direction: whoever builds AI now gets watched more closely.
The White House's pick for AI czar is likely Jay Clayton — and he may keep his day job running American intelligence at the same time. Sources told CBS News that the Trump administration has settled on Clayton, the director of national intelligence, as its likely choice to run the AI czar role, with discussions about him holding both posts concurrently. The dual-hat would be remarkable: the person shaping US AI policy would also be the person briefings on foreign intelligence capability land on. It follows Trump telling Axios last week that Clayton is "a good man" and that making him AI czar is "a good idea" when a reporter raised it, and Clayton's own line on CNBC that "super intelligence is a national security issue" — while declining, then, to confirm the job. Precise duties for the czar post remain unspecified beyond Trump's plan for a new AI Force agency, so the real story is the philosophy the appointment carries in: Clayton has argued publicly that existing consumer and product law already covers AI harms — the same line the White House delivered against the Senate's liability bill this week (we covered that argument in Hawley and Murphy: AI companies liable when agents hack). A DNI who frames superintelligence as a security file, but leans on statutes written before any of this existed, running the office meant to decide how fast the technology moves — that is the policy, in one appointment.
California just made robotaxi operators pay for blocking first responders. Governor Gavin Newsom signed Senate Bill 1246, which fines autonomous vehicle companies when their cars block police or firefighters during an emergency for more than 30 minutes, and requires them to keep first responders supplied with local, on-the-ground support when vehicles go wrong. The law also restricts remote drivers to people based in the United States holding US licenses — aimed squarely at how companies like Tesla staff the human who takes over a stuck robotaxi — and obliges operators to tell cities when vehicles fail system-wide and to station local incident technicians. The trigger was a year of visible failures: Waymo, which runs roughly 4,000 vehicles nationally with about 1,200 in the Bay Area, has blocked first responders in San Francisco 31 times since April 2025, and federal regulators sent every AV developer a letter in July demanding solutions. Waymo and Zoox say they'll comply; the rules take effect in July 2028, so the near-term test is whether the industry's voluntary fixes keep pace with a state that has decided accountability now has a price list.
What to watch: whether Clayton's confirmation fight — if he does keep both roles — turns into the first public Senate hearing on what an AI czar actually does.
Do you want the AI czar running American intelligence on the side, or is the split the point? Tell us in the comments.
