Guardrails
AI Security
AI safety
Prompt Engineering
LLMOps
NeMo-Guardrailsを試してみる
https://zenn.dev/kun432/scraps/f8dd83fb57b413
SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models
https://arxiv.org/abs/2502.12464
機密情報の流出を防ぎ、企業の安全な生成AI活用を促進する「chakoshi」のパブリックベータ版を公開
https://www.ntt.com/about-us/press-releases/news/article/2025/0219.html
生成 AI をもっと気軽に、安全に使うための「chakoshi」をリリースした話
https://engineers.ntt.com/entry/202503-about-chakoshi/entry
Exploiting Partial Compliance: The Redact-and-Recover Jailbreak
https://www.generalanalysis.com/blog/redact_and_recover