
The Security Hole at the Heart of ChatGPT and Bing
Security, Spoken
00:00
The Threat of Indirect Prompt Injections
Indirect prompt injection attacks are similar to jail breaks, a term adopted from previously breaking down the software restrictions on iPhones. Prompt injection is easier to exploit or has less requirements to be successfully exploited than other types of attacks against machine learning or AI systems. There's been a steady uptick of security researchers and technologists poking holes in LLMs.
Transcript
Play full episode