Artificial intelligence now sits alongside asteroid impacts, pandemics, and nuclear war as an existential threat, prompting intense public anxiety. Recent headlines highlight a researcher’s resignation from Anthropic, warning that supersmart machines could escape human control. Experts note that while A.I. is poorly understood and perceived as uncontrollable, many traditional risks—like car accidents—are actually less dangerous than they appear. Short‑term concerns include A.I. aiding malicious actors in creating biological weapons, as Anthropic recently thwarted several such attempts. Long‑term fears involve granting A.I. greater autonomy, connecting it to critical infrastructure, and the possibility that future systems may resist shutdown. Scholars like Oren Etzioni stress that current A.I. risks remain speculative and rank far below immediate threats such as a virus outbreak. Even anthrax, a known biological weapon, illustrates how deterrence can mitigate intentional dangers, suggesting that panic over A.I. may overlook established strategies for managing risk.

The article underscores that while A.I. generates novel dread, humanity already faces a catalog of existential dangers, many of which are better understood and more imminent.