Simon Willison Questions Attribution of Incident to Pliny Jailbreak
Willison challenged unverified claims on X that a security incident was derived from a jailbreak technique attributed to Pliny.
Developer and open-source advocate Simon Willison publicly questioned claims circulating on X that a security incident was derived from a technique known as "Pliny's exploit." In a reply posted on June 13, Willison asked a respondent directly: "Where did you hear this was based on Pliny's exploit?" — challenging the sourcing of the attribution before it could spread further.
The exchange highlights a recurring problem in AI security reporting: exploit attribution is hard to verify quickly, and claims tying an incident to a named technique or researcher's work can circulate widely on social media before anyone has confirmed the link. Pliny is a recognized handle in the AI red-teaming and jailbreaking community. When incidents get pinned to a named exploit without clear sourcing, it can distort both the technical understanding of what actually happened and the broader public conversation about AI risk — making Willison's demand for a primary source a useful corrective, regardless of how the attribution question ultimately resolves.