AI research and discussion on LessWrong. www.lesswrong.com
Measures how well Opus 4.5 as an attacker can evade prompted and fine-tuned monitors detecting data exfiltration or untrusted code execution in BashBench, finding low success against the best classifiers.
A monthly roundup of AI safety papers covering pretraining data filtering against harmful finetuning, reward hacking generalizing to misalignment, covert sandbagging against CoT monitors, safe-completions training, and probe-based defenses against cipher attacks.}
Red-teams gpt-oss-20b with a chain-of-thought repetition jailbreak, finding high robustness overall but non-monotonic attack success with repeats, CoT that contradicts final outputs, and hallucinated policies.
Shows that trusted LLM monitors in AI control protocols can be jailbroken by an untrusted model via fake conversation history, and that in-context defenses protecting the monitor substantially restore backdoor detection.