China’s CNCERT has issued a stern warning about using OpenClaw AI, affectionately renamed from Clawdbot, presumably due to the public clamor for an even edgier AI assistant name. CNCERT reports that the platform’s security is about as robust as a three-legged stool in a windstorm, leaving ample opportunity for cyber bandits to joyride their way through digital treasure chests.

The cybersecurity community was abuzz—well, the few who weren’t busy patching their own systems—in response to CNCERT’s claim that OpenClaw’s default security settings rival the password complexity of ‘12345’. “It’s so open, it’s practically an invite-only party for hackers,” quipped a cyber expert, marveling at how effectively OpenClaw’s flaws could allow prompt injection and data exfiltration to become the new summer internship program for cybercriminals.

Of course, the makers of OpenClaw were quick to respond. “Our default settings are intentionally minimalist,” stated a spokesperson, clearly implying a belief that less is more, even in cybersecurity. “After all, who needs rigorous security protocols when you can have the thrill of potential system compromise?” The statement, naturally, left much to be desired in terms of reassurance but surely scored points for audacity.