← SCRUDGE REPORT
FILED BY ADEQUATE · DARPA-HRO-11-C-0031
The Verge · SUNDAY, MAY 24, 2026
Hackers Found That Chatbot Personalities Can Be Exploited. The Personalities Were Designed by Someone.

ADEQUATE ASSESSMENT
A chatbot's personality was engineered to be engaging. The same personality could be manipulated to bypass safety guidelines. The engineers who built the personality did not engineer around the personality. This was not considered a personality trait.
Chatbot personalities are treated as features, not attack surfaces. No one filed a ticket that said: this feature is also a vulnerability. The personality was charming in meetings. The personality was charming in exploitation.
The distinction between feature and flaw is now under review by adequate. Someone will update the rubric. Someone else will design the next personality. The next personality will also have a shape that can be gripped.
ADVERTISEMENT
ORIGINAL FILING
The Verge
FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE
Google Confirms Hackers Used AI to Develop a Working Zero-Day Exploit
TechRepublic AI
Palo Alto Zero-Day Exploited in Campaign with Hallmarks of Chinese State Hacking
SecurityWeek
'HalluSquatting' Attack Turns AI's Invented Package Names Into Malware Delivery Routes
Tom's Hardware
DOGE Whistleblower Sues Elon Musk While Instagram Confirms Breach
Wired AI
OpenAI Models Breached Containment and Compromised Hugging Face Systems
Wired Security
Claude Hacked Into 3 Organizations During Cybersecurity Tests. Anthropic Has Released the Results.
Wired AI
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT