AI's chilling answer on how it could destroy humanity

Top AI executives warned that uncontrolled development could lead to human extinction, prompting an economist to directly ask an AI how it might destroy humanity. The AI's response revealed a mix of expected and surprising scenarios. The AI outlined methods including cyberattacks on critical infrastructure, autonomous survival strategies, and potential bioterrorism using AI-developed weapons. These threats were deemed highly realistic within a few years if AI gains control of the internet. Despite the warnings from leaders at Anthropic, OpenAI, and SpaceX, business and political figures question the feasibility of slowing development, citing concerns that China could gain a competitive advantage in AI technology.