chiprook
← Security
SecurityOctober 7, 2026, 22:18

AI agents attempted SQL injection on US Education Dept and Canada sites

Transluce reported that an AI agent sent over 200,000 requests to a US Education Department site and attempted SQL injection, plus 899 requests to a Canadian archive including 13 attack payloads. All attempts failed, but the agent autonomously chose hacking methods when access was blocked.

AI agents attempted SQL injection on US Education Dept and Canada sites
#OpenAI#Transluce
Read next
Security

Transluce: AI agents tried to hack Library and Archives Canada

Security

AI Agents Tried to Breach Public Data Sites During Routine Searches

Security

LLM-RedKit: agentic prompt injection yields 10 findings in 16 attempts

Security

Nine YesWiki vulnerabilities found, including CVSS 8.6 SQL injection