Jailbreaking GPT-4V via Self-Adversarial Attacks with System Prompts.

Yuanwei Wu Xiang Li Yixin Liu Pan Zhou Lichao Sun

Published in: CoRR (2023)

Keyphrases

countermeasures
security threats
watermarking scheme
malicious attacks
traffic analysis
chosen plaintext
malicious users
security mechanisms
denial of service attacks
computer security
security flaws
terrorist attacks
database
control group
cooperative
multi agent
neural network
real time