Media report · AI-ASSISTED BRIEF

Canary Experiment Finds Sensitive Data Leaking Across AI Inference Supply Chain

Report publication · YYSI publication

According to Ars Technica, Security researcher HD Moore seeded nearly two million unique, trackable secrets into AI prompts routed through inference brokers and found widespread leakage, according to Ars Technica. Findings included three hosting providers compromised by attackers who siphoned prompts, resellers logging and passing on credentials, privacy flags such as Zero Data Retention being ignored, and one Chinese AI lab fetching secret URLs from prompts.

Source & context

arstechnica.com · Read the original source ↗

This AI-assisted brief attributes information to the linked media report. It is not an official announcement or independent YYSI reporting. Consult the original report and its evidence.

Report a correction · Editorial standards