How Testers Can Break AI: Practical Techniques to Find Bias, Hallucinations, and Accessibility

As AI-powered features (especially generative AI) are rapidly integrated into modern software, testing teams face a critical challenge. Traditional testing approaches focus on correctness and performance but fail to uncover ethical risks such as bias, hallucinations, and accessibility regressions. In real projects, this has led to AI systems that technically “work” yet exclude users, generate misleading outputs, or erode trust. In this talk, Aditi addresses this gap by reframing AI quality as a testable concern and applying practical, tester-led techniques rather than data science-heavy solutions. She will demonstrate how exploratory testing, risk-based test design, prompt variation, and persona-driven scenarios can be used to intentionally stress AI behavior and reveal ethical and inclusive AI failures early in the lifecycle. She will also show how these checks can be embedded into existing test strategies and continuous testing pipelines without slowing delivery. Attendees will leave with concrete test heuristics, sample ethical AI test charters, and actionable strategies for detecting bias, hallucinations, and accessibility issues in AI-enabled systems. Most importantly, delegates will gain a repeatable approach they can immediately apply to elevate their role as quality advocates and responsible AI leaders within their teams.