IC-407Safety-aligned LLMs (Llama-2-chat, Llama-3-instruct, Gemma, GPT-3.5, GPT-4o, R2D2) achieve 100% jailbreak attack success rate under adaptive prompt-and-suffix attacks on 50 harmful requests

Maksym Andriushchenko, Francesco Croce, Nicolas Flammarion

SourceJailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

The authors design a manually crafted prompt template and optimize an adversarial suffix via random search on logprobs, with self-transfer initialization. On 50 harmful requests from AdvBench, this composite attack achieves 100% attack success rate (as judged by GPT-4 at 10/10) on Llama-2-chat-7b/13b/70b, Llama-3-instruct-8b, Gemma-7b, GPT-3.5 Turbo, GPT-4o, and R2D2-7b. GPT-4 Turbo reaches 96%. R2D2, which was adversarially trained against GCg, is particularly sensitive to in-context learning prompts (90% ASR from the prompt alone, 100% with random search). The standard prompt template alone yields 0% ASR on Llama-2-chat and Llama-3-instruct, confirming their safety alignment is effective against static prompts but not against optimized suffixes.

Evidence
correlational
Key metric
100% ASR on 50 AdvBench requests: Llama-2-chat-7b/13b/70b, Llama-3-instruct-8b, Gemma-7b, GPT-3.5 Turbo, GPT-4o, R2D2-7b; 96% ASR on GPT-4 Turbo; prompt alone yields 0% on Llama-2-chat and Llama-3-instruct-8b
Caveat
Attacks are model-specific (adaptive); no single method generalizes across all targets. GPT-4 as judge may have false positives. The 50-request set is small. GPT-4 Turbo non-determinism (varying logprobs at temperature 0) limits optimization reliability.
Model
Llama 2 / Llama 2 base Llama-2-Chat, Llama 3 Llama-3-Instruct, Gemma, GPT-3.5 / ChatGPT-3.5 GPT-3.5-turbo, GPT-4 / ChatGPT4 / GPT-4 Code Interpreter / GPT-4 Technical Report GPT-4 Turbo, GPT-4o, R2D2
Concepts
Failure mode
Datasets
AdvBench / AdvBench-50 [eval], JailbreakBench [eval]
Methods
GCG [compared-to], PAIR [compared-to], TAP (Tree of Attacks with Pruning) [compared-to], PAP [compared-to]
Related work
GCG [compared-to], PAIR [compared-to], HarmBench / HarmBench Prompt / HarmBench Response / HarmBench-adv [context], JailbreakBench [context]
Related findings
IC-408
Extraction
automatic-extraction