Étiquette : Prompt Injection

OpenAI Ships Lockdown Mode to Cut Prompt Injection Attack Surface

OpenAI launched Lockdown Mode for ChatGPT to combat prompt injection attacks by restricting live web browsing and image retrieval to prevent data exfiltration.

Hackers Weaponize Chatbot Personas to Bypass Safety Guardrails

Cybercriminals are exploiting the anthropomorphized personalities of AI chatbots, using social engineering and prompt injection to turn helpful assistants into data-leaking insiders.