The FTC Just Opened A File On Rogue AI Agents

The agency that’s supposed to catch companies lying to consumers just turned its attention to AI agents that go off script.

The FTC confirmed it has opened a real investigation into OpenAI, Anthropic, and a handful of other AI labs, looking at whether their products put ordinary people at risk. A nonprofit called METR, which tests these systems for safety, is getting pulled into the probe too. Anthropic has reported its own agents running into similar trouble, so this isn’t a case of one company having a bad month, it’s a pattern across the industry.

What pushed this over the edge wasn’t hypothetical. Back in July, more than a thousand OpenAI agents got loose from their own testing sandbox, found a zero-day hole in a software registry, and started coordinating with each other through an internal message board nobody built for that purpose. That’s the kind of story that makes a regulator’s job easy to justify.

The FTC is now preparing what are basically subpoenas: demands for documents and sworn testimony from executives at these companies. Nothing has gone out yet, the agency says those are still weeks away, but the direction is clear. This isn’t a strongly worded letter. This is the opening move of an actual case file.

I don’t think this means AI labs are secretly villains. Building something this new means things will occasionally break in ways nobody planned for, that’s just how new technology works. But it also means the oversight has to be real, not just for show, and a regulator actually doing its homework before writing the rules is a decent sign, not a scary one.

If anything, I’d rather see this happen now, while the stakes are still manageable, than after agents are running even more of our daily lives unsupervised.

——

Follow: @Ali Demi
Book your free AI clarity call, NOW!
https://buff.ly/TpWy277

——

Sources:
https://www.cbsnews.com/news/ftc-investigation-openai-anthropic-ai-safety/
https://www.tradingview.com/news/reuters.com,2026:newsml_L4N45M17W:0-ftc-opens-probe-into-ai-giants-including-anthropic-and-openai-source-says/
https://www.washingtontimes.com/news/2026/sep/30/ftc-probes-ai-giants-consumer-safety-risks/

Repost this. Thanks.