New to this? Read it in simple words
- Trump’s AI task force, the Super Intelligence Force, met for the first time on October 15.
- It set up three committees and will work for 120 days.
- It plans rules for AI labs to warn the government early about hacks and other problems.
- Those rules, and any deadlines or penalties, are not written yet.
- Readout
- A short official summary of what was said and decided at a meeting.
- Jailbreak
- A trick that makes an AI model ignore its safety rules.
- PCAST
- The President’s Council of Advisors on Science and Technology, a group of experts who advise the US president.
President Trump announced the Super Intelligence Force on October 4. It coordinates the federal government’s work on AI, which the administration now calls “Super Intelligence”. Its members met for the first time on October 15.
What the first meeting decided
The task force formally adopted its charter, Semafor reports. A White House readout, released late on October 15, says the group was formally set up for 120 days, according to Reuters.
It set up three committees. Pentagon technology chief Emil Michael leads reporting and response, and personnel chief Scott Kupor leads critical infrastructure. FTC chairman Andrew Ferguson leads law enforcement.
The effects of AI on workers and the economy will be reported to the Department of Labor, Clayton said.
From the White House readout, as Reuters and Semafor report it, and Clayton’s CNBC interview of October 16.
Early warnings from the labs
Michael’s role includes developing “protocols and mechanisms for labs to provide early warnings and systemized notifications to the government”, the readout says. These would cover breaches, hacks, jailbreaks and other problems.
A working group led by Clayton and National Cyber Director Sean Cairncross will put the White House Accord on Super Intelligence into practice. AI leaders signed that voluntary accord at the White House in September.
So far, no definition of an incident, no deadline and no penalty has been published.
A report on AI risks
The president’s science advisers, known as PCAST, will write a report on AI risks, Semafor reports. It will cover cyber security and critical infrastructure, biosecurity, and the control of models and who is accountable for them. No deadline has been given.
Any AI wrongdoing would be pursued through the FTC or the Justice Department, Clayton told CNBC.
“We set out a protocol and a path for making sure that those accords are implemented and that we’re in dialogue with the frontier labs,” Clayton said.
Sources
Every fact in this story comes from the sources below. Open them to check our work.
- 1
- 2Research · October 16, 2026US’s AI task force sets path for study and reviews, leader tells CNBC Reuters, via Yahoo News
- 3Primary source · October 16, 2026Watch CNBC’s full interview with Super Intelligence Force Chair Jay Clayton CNBC
- 4Research · October 16, 2026Trump’s AI task force starts work and tells labs to report rogue behaviour The Next Web
The White House has not published its readout of the meeting; its contents come from Reuters and Semafor, and Clayton spoke on the record to CNBC.
The task force set up committees and plans early-warning rules for AI labs, but no rules, deadlines or penalties exist yet. How we rate the mood