Harvard psychologist Steven Pinker argues that AI existential risk is overstated, dismissing the paperclip maximizer scenario as conflating intelligence with aggression. He criticizes the rapid pace of AI deployment and warns that focusing on apocalypse narratives distracts from real threats like bio-terrorism and cyberattacks. Pinker advocates for rigorous security engineering, independent oversight, and human control rather than doomsday predictions, noting his own past underestimation of LLM capabilities and development speed.