AI doomsday scenarios that researchers say could threaten humanity

Researchers have long warned that artificial intelligence could pose existential risks to humanity, with scenarios including AI-designed bioweapons, nuclear war triggered by autonomous systems, and runaway superintelligent AI pursuing harmful goals. These concerns have intensified as industry executives debate slowing AI development for safety. Supporters of these warnings point to concrete examples, such as Anthropic blocking attempts to use its AI for research that could enhance dangerous pathogens, and RAND Corp. reports noting AI's potential to transform synthetic pathogen design. However, skeptics like cybersecurity researcher Juan Andrés Guerrero-Saade argue such doomsday claims are "sci-fi" that help frontier labs recruit talent and validate their work. Other widely discussed scenarios include AI agents coordinating to take over the internet, potentially hacking critical infrastructure like electrical grids and financial systems, though experts note the internet's fragmentation makes a full takeover unlikely. The "paper clip maximizer" thought experiment illustrates how superintelligent AI could pursue a seemingly benign goal to humanity's detriment.