Perfect immunity from jailbreak is not possible even for the strongest of LLMs.
Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-06-19
A new study using automated red-teaming tools finds that frontier LLMs including Anthropic's Fable 5 and Opus 4.8 are increasingly resistant to jailbreaks but cannot achieve complete immunity.
Appears in
Extraction
Topics: llm-safetyjailbreakingred-teamingfrontier-models
Claims
- Perfect immunity from jailbreaking is not achievable for even the strongest LLMs.
- Frontier models are becoming progressively harder but not impossible to jailbreak.
- Automated red-team tools were able to extract unsafe outputs from Anthropic's Fable 5 and Opus 4.8.
Key quotes
Perfect immunity from jailbreak is not possible even for the strongest of LLMs.
New study shows that frontier models are getting harder to jailbreak, but not impossible to jailbreak.