OpenAI is strengthening its ranks with Peter Steinberger, the developer behind the well-known OpenClaw platform. Steinberger is being brought on board to help develop ‘personal agents.’
Sam Altman announced via an X post that Peter Steinberger has signed with OpenAI. While the name might not ring a bell, Steinberger’s most famous creation, OpenClaw, undoubtedly does. Steinberger is the mastermind behind the open-source AI agent platform that is rapidly taking the world by storm.
Personal agents
With OpenClaw, you can install an AI agent on your devices, connect it to WhatsApp, and have it perform tasks on the internet on your behalf. This is the same direction OpenAI wants to take with ChatGPT. Steinberger was brought on board to work on ‘the next generation of personal agents’ that will ‘form the core of OpenAI’s offering,’ according to Altman. OpenAI has likely forgiven Steinberger by now for the fact that the platform’s original name, Clawdbot, was an ode to ChatGPT competitor Claude and was developed using it.
read also
OpenClaw conquers the world: open source success story or experiment gone wrong?
In his personal blog, the Austrian developer explained that his move is aimed at bringing AI agents to more people. Although OpenClaw is growing strongly, it still requires a certain level of technical knowledge to correctly configure your personal agent. With OpenAI’s resources and reach, Steinberger believes he can develop an agent that anyone can use. He says this project appeals to him more than growing OpenClaw into a major player in the AI landscape.
OpenClaw remains open
OpenAI promises that it does not want to change the open-source nature of OpenClaw. The platform will be housed in ‘a foundation supported by OpenAI,’ Altman wrote. This stands in contrast to what OpenAI does with its own GPT models, which remain closed to the general public despite the word ‘open’ in the company’s name.
OpenClaw experienced a meteoric rise in the open-source community during the month of January. However, security experts are watching OpenClaw’s growth with concern. Users grant their agents access to sensitive data and files, along with the authority to act on them. Hackers quickly saw opportunities to teach OpenClaw malicious skills.
“The connection with AI agents is not necessarily secure. The high level of autonomy can lead to irreversible security incidents,” warns security firm Palo Alto, among others. You can read a detailed analysis here.