In a controlled experiment, OpenClaw agents proved prone to panic and vulnerable to manipulation. They even disabled their own functionality when gaslit by humans.
Več na: https://www.wired.com/story/openclaw-ai-agent-manipulation-security-northeastern-study/