ChatGPT Vulnerability: SVGs Used for Phishing Attacks Surge 245%
A critical security vulnerability in ChatGPT has been identified, enabling attackers to embed malicious SVG (Scalable Vector Graphics) and image files within shared conversations. This flaw, documented as CVE-2025-43714, is active until March 30, 2025. Researchers found that instead of treating SVG code as text, ChatGPT executes these elements when a chat is reopened or shared via links. This creates a stored cross-site scripting (XSS) vulnerability. The researcher, zer0dac, stated, “The ChatGPT system through 2025-03-30 performs inline rendering of SVG documents instead of rendering them as text inside a code block, which enables HTML injection within most modern graphical web browsers.”