ChatGPT Vulnerability Enables Phishing and Prompt Injection Attacks

Cybersecurity researchers have identified a new vulnerability, dubbed "ChatGPhish," within OpenAI's ChatGPT platform. This flaw exploits the AI assistant's inherent trust in Markdown links and images, allowing for prompt injection and sophisticated phishing attacks. The issue arises because ChatGPT's response renderer automatically processes and displays malicious Markdown content as clickable elements within its user interface.