cognitive cybersecurity intelligence

News and Analysis

Search

ChatGPT Vulnerability Lets Attackers Embed Malicious SVGs & Images in Shared Chats

ChatGPT Vulnerability Lets Attackers Embed Malicious SVGs & Images in Shared Chats

A critical vulnerability in ChatGPT (CVE-2025-43714) allows attackers to embed malicious SVG files in conversations, posing risks of phishing and harm to users. The flaw enables stored cross-site scripting (XSS) due to improper rendering of SVG code. OpenAI has disabled link-sharing as an initial mitigation, but comprehensive fixes are still pending. Caution is advised when accessing shared content.

Source: cybersecuritynews.com –

Subscribe to newsletter

Subscribe to HEAL Security Dispatch for the latest healthcare cybersecurity news and analysis.

More Posts