Superintelligence

Cited by Musk as a reason for his ongoing concern about AI risk.

Foundational text of the AI safety field; defines the vocabulary used in the AI Canon's references to alignment and Constitutional AI research.

“Completely false. It makes an assumption which Elon and some other people may have become convinced of by reading Nick Bostrom's book 'Superintelligence' or reading you know some of Eliezer Yudkowsky's writing.”

LeCun names Bostrom's Superintelligence as the source of the existential-AI-risk / hard-takeoff assumption that Musk and others adopted, and calls that core assumption 'completely false,' rejecting the book's thesis that an uncontrolled superintelligence is a likely outcome.