Scientists developed an AI monitoring agent to detect and stop harmful outputs Post author:MiamiCrypto Post published:November 20, 2023 Post category:chatGPT / Microsoft The monitoring system is designed to detect and thwart both prompt injection attacks and edge-case threats. You Might Also Like ChatGPT’s first year marked by existential fear, lawsuits and boardroom drama November 30, 2023 How to improve your coding skills using ChatGPT February 1, 2023 Meta launches community-licensed AI coding tool to the public August 24, 2023