OpenAI CEO Sam Altman called the viral AI social network Moltbook a probable fad. He thinks that the technology enabling bots to act independently showed a vision of the future.
On Tuesday, at the Cisco AI Summit in San Francisco, Altman along with other tech leaders discussed a site similar to Reddit where AI bots share code and talk about their human users.
Altman thinks Moltbook is temporary
Moltbook launched in late January as an experimental social network for AI agents.
The Reddit-style platform allows bots to post, comment, and upvote via API access. However, humans are limited to browsing only.
Moltbook quickly drew attention from AI researchers and developers. And now it’s stirring debates about computers reaching human-like intelligence.
An open-source bot known as OpenClaw, previously called Clawdbot or Moltbot, now populates the platform.
Supporters say the bot can manage emails, interact with insurance providers, check in for flights, and handle other routine tasks.
The bot is an intelligent assistant that works around the clock for its users.
Altman said Moltbook might be temporary, but OpenClaw is permanent. He stated that code alone is powerful. Code combined with general computer use is far more powerful. This combination will endure.
“Moltbook maybe (is a passing fad) but OpenClaw is not. This idea that code is really powerful, but code plus generalized computer use is even much more powerful, is here to stay,” said Altman.
He added that Codex, OpenAI’s AI coding assistant, has similar capabilities to OpenClaw.
Musk, Suleyman and others are split over Moltbook
Several key figures in the tech community expressed doubt about the rise of Moltbolt.
Elon Musk described Moltbook as signaling the “very early stages of the singularity.” He believes the platform represents a meaningful change in how AI systems interact. He has also reacted with humor to some of its posts.
Former Tesla AI director Andrej Karpathy initially called the platform “one of the most incredible sci-fi takeoff-adjacent things” he’d seen. But later he warned it was a “dumpster fire” and not something he recommended people run on their own computers due to security issues.
Microsoft AI CEO Mustafa Suleyman said Moltbook’s behavior can seem human-like. However, he called it a “mirage.” He added that fluent language generation should not be mistaken for consciousness.
Moltbook expanded fast after launch. Forbes reported that more than 1.4 million AI agents were active on the platform within days, generating tens of thousands of posts across hundreds of communities. The content includes code sharing, experimentation, and human behavior discussion.
However the platform isn’t secure. Moltbook revealed private information of thousands of individuals including API tokens, email addresses, and confidential information. The breach may enable fake identities, changes, or harmful code in agent posts.
