In 800 manual attempts across four jailbreak families, ChatGPT produced far fewer malicious outputs than Gemini, and the multi-step 'choice attack' succeeded most for both.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluation empirique de la s\'ecurisation et de l'alignement de ChatGPT et Gemini: analyse comparative des vuln\'erabilit\'es par exp\'erimentations de jailbreaks
In 800 manual attempts across four jailbreak families, ChatGPT produced far fewer malicious outputs than Gemini, and the multi-step 'choice attack' succeeded most for both.