A significant vulnerability has been discovered in OpenAI's ChatGPT, allowing attackers to exploit the AI assistant's trust in Markdown links and images to launch phishing attacks. This technique, dubbed ChatGPhish by Permiso Security, enables prompt injections, potentially deceiving users into divulging sensitive information. The vulnerability stems from the chatgpt.com response renderer's implicit trust in Markdown links and images, which can be manipulated by attackers to inject malicious prompts. As a result, ChatGPT's web summaries can be transformed into a phishing surface, putting users at risk1. The disclosure of this vulnerability highlights the security implications of large language model (LLM) developments, which can introduce new risks alongside their capabilities. This vulnerability matters to practitioners because it underscores the need for robust security measures to mitigate the risks associated with AI-powered technologies.
ChatGPhish Vulnerability Turns ChatGPT Web Summaries Into a Phishing Surface
⚡ High Priority
Why This Matters
LLM developments from OpenAI reshape both capability and risk surfaces — security implications trail the hype cycle.
References
- The Hacker News. (2026, May 29). ChatGPhish Vulnerability Turns ChatGPT Web Summaries Into a Phishing Surface. *The Hacker News*. https://thehackernews.com/2026/05/chatgphish-vulnerability-turns-chatgpt.html
Original Source
The Hacker News
Read original →