Meta acquired Moltbook, the AI agent social network that went viral because of fake posts

[highplainsdem]

Source: TechCrunch

Meta acquired Moltbook, the Reddit-like “social network” where AI agents using OpenClaw can communicate with one another. The news was first reported by Axios and later confirmed to TechCrunch.

Moltbook is joining Meta Superintelligence Labs, a Meta spokesperson told us. Moltbook creators Matt Schlicht and Ben Parr will join the team as part of the acquisition. Deal terms were not disclosed.

“The Moltbook team joining MSL opens up new ways for AI agents to work for people and businesses. Their approach to connecting agents through an always-on directory is a novel step in a rapidly developing space, and we look forward to working together to bring innovative, secure agentic experiences to everyone,” the Meta spokesperson said.

-snip-

Last month, Meta CTO Andrew Bosworth was asked about the AI agent social network in an Instagram Q&A. He said he didn’t “find it particularly interesting” that the agents talk like us, since they are trained on massive databases of human material. Rather, Bosworth was intrigued by how humans were hacking into the network, which was not a feature but a large-scale error.


Read more: https://techcrunch.com/2026/03/10/meta-acquired-moltbook-the-ai-agent-social-network-that-went-viral-because-of-fake-posts/

---

OpenClaw was created by vibe coder Peter Steinberger, whom I've seen boast about never bothering to check his AI-generated code, and who is now working for OpenAI.

AI agents like OpenClaw, and MoltBook, are tremendous security risks:

Experts are warning about VERY serious security risks with AI agents, especially Moltbook
https://www.democraticunderground.com/100220988564

[Scrivener7]

1. You can avoid the danger if you don't use Zuckerberg products.

Just as there is no risk of self inflicted 2x4 injury to your head you don't hit yourself in the head with a 2x4.

[highplainsdem]

2. Actually, if you use a friend or coworker's computer and they have an AI agent on it, that's a security

risk. As Gary Marcus warned in that OP I linked to:

https://www.democraticunderground.com/100220988564

[Scrivener7]

6. You still have to use it. So far HAL can't lock the doors if you don't use it. No telling

how long that will last though.

[Skittles]

3. so, it should fit right in

FUCK fascist FB

[highplainsdem]

4. All the AI bros are vying to have the best, most widely used AI. The one everybody thinks is

indispensable and will eventually pay a lot for, every month, through a subscription or being targeted by ads.

None of them care how much damage they do to society.

And they're all hoping to someday develop superintelligent AI, even though they know it might destroy humanity, because they want to win this race, and they hope that if AI doesn't kill all of us, it might make some of us - or at least make them - godlike and immortal.

Our tech lords are lunatics and megalomaniacs.

Did you ever see the film Mountainhead?

[Skittles]

5. haven't seen that movie, I don't do HBO either

will see if I can catch it somewhere