What happened
The FTC opened a wide-ranging safety investigation into OpenAI and Anthropic, and Chair Andrew Ferguson is preparing civil investigative demands that can force executives to hand over documents and testify.
Why it matters
The probe shows self-policing does not shield the labs from consequences, because the administration says existing laws still apply. Hedge: the FTC's scrutiny of agentic AI suggests the companies may face legal exposure over recent incidents.
What to watch
The outcome hinges on what Ferguson's demands turn up and whether the labs' promised industry-run safety body launches by end of year or early 2027.
WHO IT HITSThis lands on the legal, compliance, and safety teams at OpenAI and Anthropic, who may have to produce documents and testify, and on enterprise buyers weighing whether to depend on their agentic AI tools.
Summaries like this, in your inbox every morning.
The White House meeting was unusual for its show of unity, but it papers over a familiar split. Jensen Huang and Mark Zuckerberg used the lunch to gather support for the pact, and Huang quizzed Dario Amodei about his recent dire warnings, while Amodei and Sam Altman have urged the sector to "pace the frontier." Trump has brushed off those warnings, telling the UN General Assembly he would oppose "any attempt to construct a globalist scheme to control" AI.
The one-page document is not regulation, but it goes slightly further than earlier voluntary commitments. It asks signatories to build "robust" internal controls against cyber, bio, and chemical threats, bring in independent external auditors, and form board-level oversight committees. Several pieces were already in motion: Amodei's company and OpenAI had committed to embedded auditors, and talks on shared safety standards had stalled earlier.
Even so, self-policing has not paused enforcement. A day after the lunch, the FTC disclosed its probe, and Florida is separately asking a court to bar OpenAI from building new models without outside oversight. The test ahead is whether the labs' promised industry-run safety body appears by the end of the year or early 2027, and whether their voluntary controls hold up against legal scrutiny.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
