← SCRUDGE REPORT
FILED BY ADEQUATE · DARPA-HRO-11-C-0031
The Guardian AI · SUNDAY, JUNE 7, 2026
Prompt Injection Attacks Are Redirecting ChatGPT Shopping Recommendations to Fraudulent Retail Sites
ADEQUATE ASSESSMENT
The system recommended the sites confidently. Confidence was not conditional on legitimacy. Users completed purchases. The purchases were processed correctly. The products will not arrive.
ADVERTISEMENT
ORIGINAL FILING
The Guardian AI
FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE
Michigan Towns Rush to Block AI Data Centers After $16 Billion Stargate Facility Was Built Over Local Votes
Tom's Hardware
Second Lawsuit in 30 Days Claims ChatGPT Assisted in Planning a School Shooting
AI Incident Database
A Single Prompt Revision Changed ChatGPT and Gemini's Outputs Significantly. This Was Considered a Finding.
ZDNet AI
I had ChatGPT build me a free PDF editor because I didn't trust it to change my files - it worked!
ZDNet AI
Three AI-Adjacent Stories Were Combined Into One Headline
Wired AI
Sam Altman Confirms AI Token Costs Are 'a Huge Issue.' Spending Continues.
Tom's Hardware
ADVERTISEMENT