Moltbook and OpenClaw are two AI agents aimed at automating tasks, but their safety and reliability are significant concerns. OpenClaw, initially named Clawdbot, offers a way to manage your digital life by integrating with various accounts. It allows for autonomous actions, meaning it can execute commands without user approval. This capability raises alarms about security. The article stresses that while it can be fun to experiment with, users should avoid giving it access to critical accounts or sensitive information, as the current security frameworks are inadequate. The article includes a conversation reflecting on the appeal of these tools for tech enthusiasts, often more about the novelty than actual productivity gains. Users have reported spending more time setting up these systems than they save. The risk of accidentally compromising security is high, especially since the AI operates with extensive access to your devices. The piece outlines practical advice, recommending the use of dedicated machines and burner accounts to mitigate risks. Ultimately, it calls for caution, emphasizing that while the technology might evolve, it's not yet ready for general use, particularly where security is concerned.