Cybersecurity researchers have disclosed details of a vulnerability in OpenAI ChatGPT that leverages the artificial intelligence (AI) assistant’s implicit trust in Markdown links and images to trigger prompt injections and open the door to phishing attacks.
The technique has been codenamed ChatGPhish by Permiso Security.
“The chatgpt.com response renderer trusts Markdown links and Markdown –
Read More – The Hacker News
![AttackFeed by Joe Wagner | [Webinar] Mythos Reality Check: Beating Automated Exploitation at AI Speed - The Hacker News](https://attackfeed.com/wp-content/uploads/2026/04/miggo-webinar-h3RhI9.jpg "[Webinar] Mythos Reality Check: Beating Automated Exploitation at AI Speed")