How Cato Uses Large Language Models to Improve Data Loss Prevention

**Cato Networks DLP announcement:** Cato Networks describes a new DLP feature that uses NLP/LLM-based text embeddings (e.g., Sentence-BERT), contrastive learning, and radius-based k-NN classification to detect and block sensitive document categories (financial, legal, HR, medical, etc.), offering more context-aware detection than pattern/regex approaches and enabling advanced policy creation.