Google DeepMind Chief AI Officer: Probability of AI Causing Human Extinction "Not Zero"

TechFlow reports, August 6, according to Fortune, Google DeepMind Chief AI Readiness Officer Lila Ibrahim stated in an interview that the probability of AI causing human extinction is "not zero", and emphasized that discussing AI existential risk at the current stage is "a responsible practice". Ibrahim co-signed the "Center for AI Safety" statement with OpenAI's Sam Altman and Anthropic's Dario Amodei in 2023, listing AI extinction risk alongside nuclear war.

Regarding optimistic predictions such as Musk's "money will lose meaning by 2036" and Bezos's "millions of people will live in space by 2045", Ibrahim clearly expressed disagreement, stating that "AI is developing too fast for anyone to accurately predict the future form". She indicated that rather than envisioning future scenarios, focus should be placed on how to use AI as a tool to solve real-world problems such as extreme weather and industrial waste, and to ensure that technological dividends do not flow only to a few tech companies controlling AI.

Regarding AI extinction risk, AI godfather Geoffrey Hinton previously estimated the probability of human extinction within thirty years at 10%–20%, while Amodei believes the probability of an "extremely bad" future is about 25%.