Crypto scam triggers total ban
Peter Steinberger, the Austrian developer behind OpenClaw, has enforced a blanket no-crypto policy across the project’s Discord server.
OpenClaw is an open-source AI agent framework. It surpassed 200,000 GitHub stars since its release in late January 2026, making it one of the fastest-growing AI projects in recent memory. The ban covers any mention of Bitcoin, crypto, or related topics, even in technical contexts. A developer was removed from the server for referencing Bitcoin block height in a code discussion. Steinberger temporarily reinstated the user but maintained the rule, stating the server “enforces a no crypto mention whatsoever policy.” He added that he would never launch a coin and warned that any token claiming association with him was fraudulent.
Fake $CLAWD token sparked the crisis
The ban traces back to a chaotic rebranding in late January 2026. During the transition, scammers seized abandoned social media handles linked to the project. They used them to promote a fake Solana-based token called $CLAWD. The token surged to approximately $16 million in market capitalisation within hours. After Steinberger publicly denied any involvement, it collapsed by more than 90%. Weeks of harassment followed. The creator nearly deleted the entire project as a result.
The experience led Steinberger to implement a zero-tolerance crypto rule. He justified it as a necessary measure to protect the community from speculative financial activity and brand misuse.
Steinberger joins OpenAI as OpenClaw moves to a foundation
The crypto ban coincides with a major shift in Steinberger’s professional trajectory. He announced he is joining OpenAI to work on bringing agents to everyone. He stated that OpenClaw “will move to a foundation and stay open and independent.” OpenAI’s acquisition of Steinberger is widely seen as a significant blow to Anthropic. OpenClaw previously defaulted to Anthropic’s Claude models.
Analysts expect it to switch to GPT models following the move. Meta has separately banned the use of OpenClaw on work devices over security and data protection concerns. Startups, however, are embracing it cautiously within isolated virtual machines. The diverging corporate approaches highlight the uncertainty surrounding AI agent tools as they mature.
