How Is Predictive AI Shaping Our World? With AI Philosopher Carissa Véliz

In this episode of Intelligence Squared, philosopher Carissa Véliz joins host Tom Chatfield to explore the profound societal implications of predictive AI, drawing parallels between ancient oracles and today's algorithmic forecasting systems. Véliz argues that predictions—often mistaken for facts—are not neutral descriptions but powerful speech acts that shape behavior and wield influence. She warns that when institutions and individuals uncritically accept bold, confident predictions about the future, they surrender agency and democratic control to those who make them. The conversation traces the historical roots of prediction from Delphi to modern AI, revealing how prediction has always been entangled with power, manipulation, and ideology. Véliz critiques the overreliance on quantified, long-term forecasting—especially in movements like effective altruism—highlighting how such approaches ignore epistemic humility and can lead to unintended harm. She champions a more virtuous, open-minded, and playful approach to the future: one grounded in curiosity, empathy, and serendipity rather than deterministic prediction. Ultimately, the episode calls for a radical rethinking of how we engage with prediction—not as a tool to discover a fixed future, but as a means to co-create a better one through ethical imagination and collective action.