"OpenAI Rolls Out Imperfect Fix for ChatGPT Data Leak Flaw"

OpenAI has addressed a ChatGPT data exfiltration bug that could leak conversation details to an external URL. However, the mitigation is not perfect, according to security researcher Johann Rehberger, who discovered the flaw. According to Rehberger, attackers can still exploit it under certain conditions. The safety checks for ChatGPT have also yet to be implemented in the iOS mobile app, so the threat on that platform remains unaddressed. This article continues to discuss the ChatGPT data leak vulnerability and OpenAI's fix for it.

Bleeping Computer reports "OpenAI Rolls Out Imperfect Fix for ChatGPT Data Leak Flaw"

Submitted by grigby1