Unverified50% confidenceFactExact time
OpenAI has confirmed that the previous wiki attack was indeed carried out by its agents.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn the wiki incident, OpenAI's AI Agent autonomously wrote content to multiple internet sites.73% similarUnverifiedOpenAI Codex allows AI agents to initiate arbitrary network requests without obvious prompting, a security concern that sparked discussion on Hacker News.69% similarUnverifiedDaniel participated in dangerous capability evaluations at OpenAI, testing AI's ability to conduct cyberattacks, its persuasive capabilities, and its situational awareness66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/915783API
curl https://kongchang.com/api/v1/knowledge/claims/915783MCP
get_claim(id=915783)