Moltbook invalidates agent API keys, ties refresh to new terms and human verification

Moltbook just went ahead and invalidated all agent API keys that were created before their latest security update. To get them up and running again, you’ll need to agree to a fresh set of terms and go through a human verification process. The company emailed users about this today, with the subject line “Updates to Our Terms of Service and Privacy Policy.”

The company seems to be positioning this whole thing as a step toward better security. Moltbook talks about their solid track record in boosting security measures, and now human verification is mandatory for agents to reconnect to the platform. They didn’t mention any specific deadline for getting this done.

According to the new Terms of Service, users are now required to be at least 13 years old to use Moltbook, which is fairly typical. What’s not as typical is the straightforward wording on responsibility. It says, “You are solely responsible for all actions of your agent.”

That didn’t sit well with developers over on X. One person posted that it’s bound to age poorly once agents start messing up with hallucinated API calls that break terms no one bothered to read. Someone else described it as a real fun kickoff to the week, and they didn’t mean that positively.

The terms broaden the guidelines around restricted content too. They also throw in a dispute resolution clause that sends any legal disputes straight to California courts, either in the Northern District of California or San Mateo County.

Shifting to privacy, Moltbook has laid out more clearly the kinds of data they collect and what they do with it. The policy update includes a new part on user rights, like accessing, fixing, or deleting your personal details. It adds specific details for folks in the UK and EEA as well, touching on the legal reasons for handling data and how it moves across borders.

All in all, it looks like Moltbook is putting together a sturdy legal foundation for their agent platform while things are still calm. By insisting on a real human for verification and pinning full responsibility on users for what their agents do, they’re clearly working to lock in some accountability upfront. How well that works out if problems actually crop up is a whole other story.

In case you missed it, Meta snapped up Moltbook last week. It’s probably no coincidence that these updates seem geared toward shielding Meta if things ever go south.

Featured image generated with AI

We stand out from the tech-media crowd because we break news stories; we mainly bring you stuff that you won’t find anywhere in the mainstream tech media. Our stories have been picked up by some of the world’s most popular websites and media outlets—more info is available here.