AI News · #alignment · anthropic-research

Anthropic Research · 2026-07-13 제목번역

Claude의 가치가 모델과 언어에 따라 어떻게 다른지

How Claude's values vary by model and language

#claude #alignment #multilingual #ai-values #model-variation

Anthropic Research · 2026-07-08 제목번역

Alignment faking in large language models

#large-language-models #safety #alignment #llms #deception

Anthropic Research · 2026-07-08 제목번역

Constitutional Classifiers: Defending against universal jailbreaks

#ai-safety #alignment #constitutional-ai #jailbreak-defense