# Researchers Expose OpenAI Vulnerability Using Rival AI Model

> Source: <https://osintsights.com/researchers-expose-openai-vulnerability-using-rival-ai-model>
> Published: 2026-09-18 18:03:04+00:00

Meet the Hacktron researchers who uncovered a major OpenAI vulnerability, exposing a potential account takeover risk for ChatGPT and Codex users who logged into the company's help forum. They cleverly used a rival AI model to develop an exploit, chaining flaws to gain access to multiple OpenAI accounts.