Validating and testing RAG systems to manage generative AI risks in real-world use.

Cohort 1

Software Systems Architect

Richard McKnight

Learn to validate and govern RAG-based generative AI with cutting-edge human-calibrated testing and NIST AI 600-1 guidance.

Patrick Hall

Principal Scientist at HallResearch.ai, GWU Faculty

Agus Sudjianto

Senior VP, H2o.ai and Former EVP & Head of Model Risk, Wells Fargo

Governing GenAI - From Prompts to Policy

Hall Research AI

<h3>Why I enrolled</h3>I want to cultivate a deep understanding of the risks associated with GenAI in the context of the rewards and benefits of GenAI.I want to be able to transform that understanding into the capability to:<ul class="list-disc ml-8"><li>develop strategies, and create specific plans for managing and mitigating the risk</li><li>successfully execute those plans</li></ul>I want to be able to measure the success of these executed plansI want to be able to do this across multiple domainsThrough well thought out and executed plans I want to be able to:<ul class="list-disc ml-8"><li>address the concerns of tentative/nervous adopters</li><li>cultivate a healthy respect for risk among the "AI can do anything and will save the world" adopters</li></ul>I want to be able to leave projects with a plan for continuous evaluation.<h3>What I’ll apply</h3>I can see some immediate benefits .....<ul class="list-disc ml-8"><li>Representing Content as Knowledge graphs</li><li>Taking a structured approach to validating the performance of an LLM/RAG based system</li></ul>Additionally, the ability to articulate the validation process will help me:<ul class="list-disc ml-8"><li>address the concerns of tentative/nervous adopters</li><li>cultivate a healthy respect for risk among the "AI can do anything and will save the world" adopters</li></ul>

Certificate

About this course