Doomsday AI in a decade? Experts debate if superintelligence could end humanity

AI safety researchers and industry voices warn that a self-improving AI could pose an existential threat within ten years, outlining scenarios from autonomous weapons and viral design to systems that act without human control; many experts argue the predictions are speculative and lack verifiable evidence, but they urge precautionary governance and safety measures to prevent misuse.