News on Medial

Microsoft and OpenAI say hackers are using ChatGPT to improve cyberattacks

The Verge · 1y

Microsoft and OpenAI have identified instances of hackers using large language models (LLMs) like ChatGPT to enhance their cyberattacks. These hackers, including groups from Russia, North Korea, Iran, and China, utilize LLMs to refine attack scripts, develop social engineering techniques, and research potential targets. While there have been no significant attacks observed yet, Microsoft and OpenAI are actively shutting down associated accounts and assets. They also express concerns about future uses, such as voice impersonation, and stress the importance of using AI to defend against AI attacks.