Indian-origin techies hack OpenAI using its rival Claude in 72 hrs

Three Indian-origin researchers said they used Anthropic's Claude in July to access an OpenAI employee's ChatGPT account and reach the company’s internal GitHub environment. “It took us less than 72 hours,” they said, adding that OpenAI paid them $6,500 bug bounty. They also used the compromised account’s access to Codex to create a harmless pull request in OpenAI's internal monorepo.