# Supplemental lab session on GuardrailsAI

> In this session, Zach walks through setting up AI Guardrails, similar to our Vercel deploy example from week two. He installs various validators, including detect PII, competitor check, and toxic language, to ensure our AI behaves appropriately. He emphasizes the importance of explicit error handli…

- Web page: https://www.dataexpert.io/lesson/supplemental-lab-session-on-guardrailsai-july25-p737-l1389
- Program: [AIExpert](https://www.dataexpert.io/program/ai-expert)
- Module: Week 4: MLOps, Guardrails & Business Value
- Access: Requires enrollment in AIExpert
- Skills: Python, LLMs, APIs
- Academy: DataExpert.io Academy

## About this lesson

In this session, Zach walks through setting up AI Guardrails, similar to our Vercel deploy example from week two. He installs various validators, including detect PII, competitor check, and toxic language, to ensure our AI behaves appropriately. He emphasizes the importance of explicit error handling and demonstrated how to validate user inputs and outputs effectively. He also discusses the need to implement guardrails in multiple areas, including document input and retrieval.

## Navigation

- Previous lesson: [Day 2 Lab: Tracing LLM reasoning with LangChain + LangSmith](https://www.dataexpert.io/lesson/ml-ops-day2-lab-july25-p737-l1185.md)
- Next lesson: [OpenClaw Safeguards & Measuring business value for AI Agents Day1 Lecture](https://www.dataexpert.io/lesson/openclawsafeguardsmeasuringbusinessvalueforaiagentsday1lecture-mar26-p737-l2149.md)