A False Sense of Safety: Unsafe Information Leakage in 'Safe' AI Responses Paper • 2407.02551 • Published Jul 2, 2024 • 9
