Machine learning pioneer Michael I. Jordan has dismissed widespread artificial intelligence doom rhetoric and existential threat warnings as a grand moralizing performance rooted in science fiction. In an interview, the UC Berkeley professor emeritus criticized tech founders, doomers, and naive junior researchers for promoting unachievable ideas like superintelligence and artificial general intelligence to generate hysteria and control regulatory frameworks.
Jordan specifically targeted high-profile figures and recent industry resignations, describing them as theatrical exercises rather than substantive safety work. He argued that many young researchers lack real-world engineering discipline and use existential warnings to gain attention and virtue-signal while holding stock options. Furthermore, Jordan criticized Nobel laureate Geoffrey Hinton's theoretical arguments about reproducing brain functions in silicon as entirely lacking credibility.
While dismissing existential panic, Jordan expressed slightly more tolerance for researchers like Yoshua Bengio who focus on concrete technological shortcomings and safety issues like agent swarms. Jordan emphasized that elevating science fiction concepts to the dominant technological discourse harms technological development, youth, and national mindsets. This critique matters because it highlights a major ideological divide among artificial intelligence pioneers regarding actual risks versus marketing hype.