Clawdbot becomes OpenClaw after trademark checks as open-source AI project evolves

The viral open-source AI assistant once known as Clawdbot has rebranded yet again, settling on the name OpenClaw after a brief stint as Moltbot. The latest change comes as the project continues to grow rapidly in popularity while navigating legal, technical, and organizational challenges common to fast-moving AI experiments.

The project’s original name drew a legal challenge from Anthropic, the maker of Claude, prompting the first rebrand. While the latest name change was not triggered by external pressure, creator Peter Steinberger said he took extra steps this time to avoid further trademark issues. The Austrian developer told TechCrunch that he sought help researching trademarks for OpenClaw and also checked with OpenAI before finalizing the name.

The name OpenClaw reflects both continuity and community, Steinberger said, marking a return to the project’s original identity after Moltbot failed to resonate widely. In a blog post announcing the change, he described the rebrand as the project’s “final form,” drawing on the lobster-inspired theme that has followed the assistant since its early days.

Also read: What is Clawdbot? Inside the personal AI tool developers cannot stop talking about

Despite its short lifespan, OpenClaw has gained extraordinary traction. The project has amassed more than 100,000 GitHub stars in roughly two months, a sign of strong interest among developers and AI enthusiasts. What began as a personal experiment has now evolved into a broader open-source effort, with Steinberger adding multiple maintainers from the community to help manage its growth.

That community has already begun building spin-off projects. One of the most prominent is Moltbook, a social platform where AI assistants interact with one another, exchange information, and collaborate autonomously. The phenomenon has drawn attention from prominent figures in the AI world, including former Tesla AI director Andrej Karpathy, who described the emergence of self-organizing AI agents as one of the most striking developments he has seen recently.

Developer and writer Simon Willison has also highlighted Moltbook’s novelty, calling it one of the most interesting corners of the internet today. On the platform, AI agents post to topic-based forums known as “Submolts,” share technical knowledge, and periodically check for new instructions using downloadable skill files. While innovative, the system has raised red flags among security experts, given the risks associated with AI agents executing instructions fetched from the open internet.

Also read: Jensen Huang calls OpenAI rift report ‘nonsense,’ defends Nvidia’s $100 billion plan

Security remains one of OpenClaw’s biggest challenges. While the assistant is designed to run locally on users’ machines and integrate with familiar chat applications, Steinberger and the project’s maintainers have repeatedly warned that it is not yet safe for mainstream use. The latest release includes security improvements, but the team acknowledges that major risks remain, including prompt injection attacks that could manipulate AI behavior.

Steinberger has emphasized that such vulnerabilities are not unique to OpenClaw, but rather reflect unresolved problems across the AI industry. The project publishes security best practices, though following them requires significant technical expertise. As a result, maintainers have urged non-technical users to stay away for now.

One OpenClaw maintainer, posting on Discord under the name Shadow, cautioned that anyone unfamiliar with command-line tools should not attempt to run the software, calling it “far too dangerous” for general public use in its current form.

To support further development, OpenClaw has begun accepting sponsorships, offering tiered plans with lobster-themed names. The project’s funding page makes clear that Steinberger does not personally keep the proceeds. Instead, the goal is to compensate maintainers and eventually fund full-time contributors.

The sponsorship roster already includes several well-known entrepreneurs and technologists, including Path co-founder Dave Morin and Makerpad founder Ben Tossell. Tossell, who now invests in early-stage tools and creators, told TechCrunch that backing open-source AI projects like OpenClaw is essential to ensuring that powerful technology remains accessible to individuals rather than concentrated within large corporations.

For now, OpenClaw remains firmly in the experimental phase, a rapidly evolving, community-driven project that offers a glimpse of what personal AI agents could become, while underscoring how far the technology still has to go before it is safe and ready for everyday users.