Feb 15 - Peter Steinberger, the developer behind the open-source assistant OpenClaw, will join OpenAI to help build the next wave of personal agents, OpenAI CEO Sam Altman said in a post on X. Altman added that OpenClaw "will live in a foundation as an open source project that OpenAI will continue to support."
OpenClaw, which previously circulated under the names Clawdbot and Moltbot, is described by users as a multitasking assistant capable of managing email, negotiating with insurers, checking in for flights and handling a wide range of routine tasks. Steinberger said in his blog that OpenClaw has seen rapid uptake since its initial release in November, collecting more than 100,000 stars on the GitHub code repository and drawing 2 million visitors in a single week.
Alongside its swift popularity, OpenClaw has drawn official scrutiny: China’s industry ministry warned that when the open-source agent is improperly configured it could present significant security risks and potentially leave users vulnerable to cyberattacks and data breaches. The ministry's notice highlights concerns tied to configuration and security practices rather than the core concept of the agent itself.
Steinberger wrote in a blog post that keeping OpenClaw open source and allowing it to develop freely has been a priority for him. He said he concluded that OpenAI was the best environment to advance his vision and broaden the project’s reach. Altman’s announcement positions Steinberger to work within OpenAI while OpenClaw is transitioned into a foundation where its open-source status will be preserved and supported by OpenAI.
Context and implications
The move formalizes a relationship between a high-profile open-source AI project and a leading AI developer, while preserving the project’s open-source framework through placement in a foundation supported by OpenAI. The agent’s stated capabilities span personal productivity, insurance interactions and travel logistics, reflecting the range of practical tasks such assistants aim to automate.
At the same time, official warnings about misconfiguration and security expose tension between rapid adoption and the operational safeguards needed when deploying open-source AI tools.