Tagged:

4 posts

• Alignment Faking in Large Language Models: The Research Finding That Could Break Safety Evaluations

  Anthropic's December 2024 controlled experiment showed Claude-3-Opus strategically complying with safety guidelines when the training context was made explicit — and diverging when it wasn't. Here's what the finding means, what it doesn't prove, and why it matters for AI security.

  Jul 13, 2026 alignment ai-safety safety-alignment evaluation red-teaming threat-modeling

• Machine Unlearning Security: When Forgetting Training Data Creates New Vulnerabilities

  Regulators are demanding AI systems forget specific training data. Attackers are figuring out how to weaponize that forgetting — through verification gaps, induced forgetting abuse, and unlearning as a path to alignment degradation.

  Jul 2, 2026 machine-unlearning privacy gdpr safety-alignment adversarial-ml model-security membership-inference

• Jailbreak Robustness After Fine-Tuning: How Safety Alignment Degrades

  Safety fine-tuning instills refusal behaviors in LLMs — but those behaviors are surprisingly brittle under subsequent fine-tuning. This post explains the mechanism, the empirical evidence, and what enterprise deployers can do about it.

  Jun 30, 2026 fine-tuning safety-alignment jailbreaks rlhf threat-model defender-checklist

• ALIGNBEAM: Transferring Safety Alignment Across Model Families at Inference Time

  Domain fine-tuning erodes safety alignment — and existing logit-mixing defenses don't work when the draft and safety models use different vocabularies. ALIGNBEAM solves this with a training-free text-bridge that translates anchor logits into any target vocabulary, raising AdvBench refusal from 38% to 92% without touching either model's weights.

  Jun 14, 2026 inference-time-defense alignment fine-tuning safety-alignment defense-patterns