AiGovOps for Testers: Validating AI Systems You Can Ship — and Defend

AI is in the systems you're testing — and the rules just changed. With the EU AI Act in full enforcement, 250+ U.S. state bills in motion, and audit demands rising, "we tested it" is no longer enough. Regulators, customers, and your own legal team want evidence: who validated what, against which risks, with what controls running when the model shipped. The good news: testers are the natural front line for AI governance. You already own validation, traceability, and the test-evidence chain that auditors will ask for. The opportunity is to make that work load-bearing for AI systems — and to do it at delivery speed, not policy speed. Ken Johnston walks you through AiGovOps: the practice of implementing AI governance as code, embedded directly in test plans, pipelines, and observability. Through hands-on exercises, you'll unpack real-world AI HARMS and introduce the HARMS framework for identifying where AI systems break user trust before they reach production. Test patterns for probabilistic and non-deterministic outputs — drift, hallucination, bias, prompt injection, jailbreak resistance. Guardrails, human-in-the-loop checkpoints, and continuous validation as runtime testing, not policy theater. The audit-ready evidence trail your test results should already be producing. This isn't about adding a slow review board to your pipeline. It's about making your existing test discipline the audit-ready governance layer your organization already needs. This tutorial is not about building AI models. It's about running and validating them in production without shipping harm, fines, or a career-defining incident. Attendees leave with a working AiGovOps first 30 days engagement plan, the HARMS worksheet, and concrete patterns they can apply to the AI features already in their backlog.