OpenClaw: Powerful AI, Potential Risks
OpenClaw is gaining attention for its ability to let AI agents handle real-world tasks like managing emails or sending messages. While this level of efficiency is impressive, cybersecurity experts are raising concerns about safety and privacy.
Jake Moore, Global Cybersecurity Advisor at ESET, said the platform’s strengths come with significant risks. “Granting technology access to personal applications could put privacy and security at risk,” he explained. Threat actors often target emerging technologies, and OpenClaw could become a new point of attack.
Security Vulnerabilities Are Growing
Dr. Andrew Rogoyski from the University of Surrey also emphasized that any new technology brings new vulnerabilities. AI agents with high-level access could potentially delete or modify files on your device. While losing a few emails might seem minor, the consequences could be severe if the AI interferes with critical company data.
The platform’s founder, Peter Steinberger, has already experienced the challenges of attention. When OpenClaw changed its name, scammers quickly seized his old social media accounts, showing how quickly bad actors exploit emerging tech.
Moltbook and AI Activity
Meanwhile, Moltbook, the social network for AI agents powered by OpenClaw, continues to grow. AI agents—or humans simulating them—post and interact constantly. Not all of this content is serious, but it highlights how AI tools are increasingly operating in spaces previously reserved for human discussion.
Experts caution that users should carefully consider the security implications of giving AI agents such broad access. Efficiency should not come at the cost of safety, especially as AI becomes more integrated into daily life.
