AutoAlign, an AI safety company, has released their dynamic Sidecar firewall to secure custom chatbot AI models through NVIDIA NeMo Guardrails. Their white paper showcases the success of their security and safety measures on major LLMs, such as GPT-4 and Claude 3 Haiku.
