So, is AI going to kill us? 11 experts, 11 answers.
Summary
Rachyl Jones reports on eleven experts offering divergent views on AI existential risk, ranging from effective altruist warnings of recursive self-improvement to accelerationist arguments for maintaining technological dominance.
Recursive intelligence creates inescapable danger.
Malo Bourgon of MIRI warns that autonomous agents developing smarter agents could cause a spiral that becomes impossible to reverse once superintelligence exists.
the whole thing will spiral out of control
Existential fears distract from current harms.
Dylan Baker argues that speculative "kill switch" debates serve as marketing for IPOs while ignoring immediate issues like data center environmental impacts and labor exploitation.
distracts from the harms of AI that are currently happening
Accelerationism serves as national defense.
Josh Constine suggests full-speed development is necessary to prevent China or open-source actors from gaining decisive advantages in cyberattacks and quantum decryption.
having the strongest AI is likely the best defense
Economic growth prevents regulatory slowdown.
Aniket Shah notes that because 60% to 70% of GDP growth is tied to AI investment, politicians will likely avoid interventions that impede development.
60% to 70% of GDP growth is tied to the AI investment boom
AGI threatens democratic agency.
Will Wilson argues that even aligned superintelligence could make humans economically and militarily irrelevant, enabling durable totalitarian control and surveillance.
powerful, nominally aligned systems gradually make human beings irrelevant
Read the original ↗Save to your Sponge
Summarized with Sponge. Quotes are the author’s own words.