AI is reshaping every step of the experimentation lifecycle — from generating hypotheses to interpreting results and turning them into actionable recommendations. What used to take days of back-and-forth between data teams, product managers, and engineers is increasingly something an AI agent can orchestrate end-to-end. In this session, we explore how AI is automating the repetitive parts of testing, dramatically lowering the bar for teams to run high-quality experiments at scale, and shifting the data analyst's role from test executor to learning architect. We'll look at where are the most experimentation programs bottlenecked and what AI has to offer there.