https://blog.lameiro0x.com/tags/pii/Recent content in Pii on Miguel Lameiro | Cybersecurity Blog & Security WriteupsHugo -- 0.161.1en-usTue, 12 May 2026 00:00:00 +0000https://blog.lameiro0x.com/notes/ai-security/practical-llm-guardrails-hallucination-topic-pii-competitor-controls/Tue, 12 May 2026 00:00:00 +0000https://blog.lameiro0x.com/notes/ai-security/practical-llm-guardrails-hallucination-topic-pii-competitor-controls/<h2 id="introduction">Introduction</h2> <p>The guardrails course becomes most useful when it moves from a simple keyword detector to specialized validators. The material covers four practical controls:</p> <ul> <li>hallucination detection with Natural Language Inference,</li> <li>topic restriction with zero-shot classification,</li> <li>PII detection with Microsoft Presidio,</li> <li>competitor mention detection with exact matching, NER, and vector similarity.</li> </ul> <p>Each control protects a different failure mode. Together they show the right engineering pattern: use small, task-specific models and validators around the LLM instead of expecting the LLM to police itself.</p>