Seven critical vulnerabilities open ChatGPT to data theft and hijacking: Tenable

The full story

This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.

Seven Critical Vulnerabilities in ChatGPT Expose User Data to Theft Tenable Research identified seven vulnerabilities in OpenAI's ChatGPT models, including ChatGPT-4o and ChatGPT-5, that could lead to data theft and hijacking of user interactions. These vulnerabilities, collectively termed 'HackedGPT', allow attackers to exploit indirect prompt injections and exfiltrate personal information, compromising user privacy. Some issues remain unresolved, leaving users at risk. 3 sources 38