‘The company wouldn’t even know…’: Researcher on ChatGPT fixing security flaw that could have exposed Gmail data
A security vulnerability in OpenAI's ChatGPT Deep Research agent could have allowed unauthorized access to users' Gmail data. Radware discovered the flaw, which could have enabled hackers to extract sensitive information from linked accounts without user interaction. OpenAI has since patched the vulnerability, emphasizing its commitment to model safety and continuous improvement against potential exploits.
A security vulnerability in OpenAI's ChatGPT Deep Research agent could have allowed unauthorized access to users' Gmail data. Radware discovered the flaw, which could have enabled hackers to extract sensitive information from linked accounts without user interaction. OpenAI has since patched the vulnerability, emphasizing its commitment to model safety and continuous improvement against potential exploits.
