AI is rapidly changing the way we approach software testing. But let’s be honest, many teams are rushing into AI driven automation without truly understanding its limitations, resulting in wasted budgets, unreliable test results, and frustrated teams. In this session, I’ll share the real story of how our testing team nearly lost control when AI-based automation started making the wrong decisions, misclassifying failures, and confusing false positives with real defects. 

What did we do? We didn’t throw AI away. Instead, we developed a strategy to harness its strengths while keeping human intelligence in control. This talk will focus on when to trust AI, when to question it, and when to override it. 

You’ll hear practical lessons from our experience on how we optimized test case selection, tuned AI models, and prevented automation from becoming a money pit. Whether you’re an AI enthusiast or a skeptic, this session will challenge your assumptions and help you prepare for the next era of testing.