Skip to content

OpenAI hires Peter Steinberger, creator of OpenClaw

Person using a robotic arm with a blue card next to a laptop displaying code on a wooden desk in an office.

OpenAI has made a major move in the talent market by hiring Peter Steinberger, the creator of the viral OpenClaw project. By bringing this exceptional developer into the company, Sam Altman’s business is signalling a clear objective: turn ChatGPT into a genuine personal agent able to take practical action on our behalf.

Peter Steinberger joins OpenAI

This is a genuine shockwave for Silicon Valley. Peter Steinberger, whose work has recently set the technology community alight, has officially joined OpenAI. For its chief executive, Sam Altman, the appointment is an obvious strategic decision. He has been full of praise for the new hire: “He is a genius who has a multitude of incredible ideas about the future of very intelligent agents interacting with one another to carry out very useful tasks for people”.

OpenClaw: an assistant that takes action

Until now, Steinberger had led the OpenClaw project on his own. Having become a GitHub star with more than 100,000 stars, the initiative puts forward a groundbreaking concept: an assistant that can go deep into your system to manage your emails, book your flights or speak to your insurer. Unlike conventional chatbots, which merely hold conversations, Steinberger’s tool is designed to actually get things done.

Serious security flaws

OpenClaw’s rise has not, however, been straightforward. The project was forced to change its name several times, moving from Clawdbot to Moltbot before becoming OpenClaw, mainly because of a trademark dispute with Anthropic. That uncertainty was exploited by scammers attempting to launch fake cryptocurrencies using the project’s image.

This creates a cybersecurity nightmare. Because the system requires extensive access to a PC and personal accounts, thousands of poorly configured dashboards ended up exposed online. That is not all: researchers also found credentials and access keys being shared in plain text, potentially enabling large-scale hacking.

My next mission is to create an agent that even my mother could use. This will require much bigger changes, careful thought about how to do it safely, as well as access to the latest models and research,” Steinberger acknowledges in a blog post.

Competing in AI agents

The appointment comes as OpenAI clearly steps up its efforts in the AI agents space. Amid fierce competition from Google and Anthropic, the company intends to establish itself as a leader in this area, widely seen as one of the most competitive battlegrounds in the AI ecosystem.

As for OpenClaw’s future, users can be reassured. The project will not be absorbed or shut down: it will become an independent foundation and remain open source. OpenAI has committed to financially supporting the structure, allowing the community to keep developing the assistant while benefiting from its creator’s expertise.

Comments

No comments yet. Be the first to comment!

Leave a Comment