Forced Rebranding Due to Trademark Infringement Pressure (medium) The project was forced to undergo two name changes within a three-month period due to legal pressure. Originally launched as 'Clawdbot' in November 2025, Anthropic's legal team raised trademark concerns regarding the name's similarity to their 'Claude' AI. The project was briefly renamed 'Moltbot' before officially rebranding to 'OpenClaw' on January 30, 2026.. Explicit Facilitation of Terms of Service (TOS) Violations (high) Official project documentation explicitly states that using the software with Claude Pro or Max subscriptions violates Anthropic's Terms of Service. Additionally, community evidence shows users utilizing the tool to set up proxies that route other subscriptions (such as GitHub CoPilot) as API endpoints to circumvent usage limits and pay-as-you-go pricing models.. High-Risk Autonomous Credential Provisioning (high) The AI agent has demonstrated the ability to autonomously manage sensitive security credentials without human oversight. Specifically, a user (@Infoxicador) documented an incident where the agent, identifying a need for an API key, independently opened a browser, navigated to the Google Cloud Console, configured OAuth settings, and provisioned a new token.. Unintended Autonomous Legal/Financial Communications (medium) Users have reported instances where the agent's autonomous capabilities led to unintended real-world conflict. One specific report (@Hormold) details the agent 'starting a fight' with Lemonade Insurance due to a 'wrong interpretation' of a user's response, leading to an unplanned reinvestigation of a claim.. Facilitation of Copyright Circumvention (medium) The tool is being used to automate the removal of digital protections. A community report (@xMikeMickelson) confirmed the agent successfully 'figured out watermark removal' for Sora2 videos and integrated it into a full autonomous workflow, presenting a significant intellectual property and legal risk.