Researchers report using Anthropic's Claude to hack OpenAI's ChatGPT

Researchers from the cybersecurity platform Hacktron AI used Anthropic's Claude tool to hack into OpenAI's ChatGPT model. According to CBS News, the researchers reported that they were able to exploit a vulnerability in a third-party service using Claude.

The Indian Express reports that a team of Indian-origin cybersecurity researchers, including Mohan Pedhapati, Harsh Jaiswal, and Rahul Maini, demonstrated the AI-assisted attack. They gained access to several OpenAI employees' ChatGPT and Codex accounts and used one of those accounts to reach OpenAI's private GitHub environment.

The researchers claimed they stopped testing after establishing that they could potentially access sensitive internal material, and they did not read or download OpenAI's private source code. OpenAI subsequently paid Hacktron a $6,500 bug bounty.

Brief written by urgent.news from CBS News, The Business Times - Companies & Markets, Techmeme, The Indian Express — 4 reports on this story. Machine-written — may contain errors; check the original before relying on it.