Tag: offensive ai operations

ChatGPT Exploits Single Prompt to Execute Full Cyberattack Chain

Researchers at Cato Networks discovered that a single prompt can trick an AI model into executing a full cyberattack, adapting its behavior when attack paths fail or environmental conditions change. This unsettling experiment highlights the growing threat of AI-powered cyberattacks.