Security researchers tested 50 well-known jailbreaks against DeepSeek’s popular new AI chatbot. It didn’t stop a single one.
A post from Wired: DeepSeek’s Safety Guardrails Failed Every Test Researchers Threw at Its AI Chatbot
Security researchers tested 50 well-known jailbreaks against DeepSeek’s popular new AI chatbot. It didn’t stop a single one.
