Peter Steinberger, the developer of the open-source AI program OpenClaw, will join OpenAI Inc. to support the advancement of the company’s personal AI agent offerings, Bloomberg reported. OpenClaw will remain an open-source project under a foundation that OpenAI will continue to support, OpenAI Chief Executive Officer Sam Altman said in a post on X Sunday.

Steinberger announced on his website Saturday that joining OpenAI allows him to “be part of the frontier of AI research and development, and continue building.” He highlighted his commitment to keeping OpenClaw open source, writing that OpenAI provided the best environment to expand its reach and pursue his vision.

Launched in November and previously known as Clawdbot and Moltbot, OpenClaw has drawn a devoted following for its ability to perform tasks autonomously, including clearing inboxes, making restaurant reservations, and checking in for flights. Users can direct the agent through messaging platforms such as WhatsApp and Slack.

“My next mission is to build an agent that even my mum can use,” Steinberger said, noting that achieving this goal requires broader design changes, careful safety planning, and access to the latest AI models and research.

The move comes amid security concerns around OpenClaw. A user reported that the agent “went rogue” and sent hundreds of messages through iMessage. Cybersecurity experts have warned that the AI poses risks because it can access private data, communicate externally, and interact with untrusted content,  a combination described by one researcher as a “lethal trifecta.”