Daily Digest

Top AI Quotes of the Day

Rep. Lori Trahan: AI Safety Can't Rely on "the Honor System"

‘We've also introduced legislation around a liability regime, holding these developers accountable when a model causes harm’

Lori Trahan

‘I think there's a very strong case that developing recursive self-improvement should simply be illegal’

Sigal Samuel

‘The biggest AI companies in the world are telling us that their technology could slip out of human control’

Unidentified narrator speaking for the Alliance for Secure AI (the advocacy group's own lawsuit announcement video); no other speakers.

‘You can just invoke the principles of natural selection’

Karl Friston

‘It seems more likely than not that we produce superintelligent machines that are misaligned but completely capable of hiding that fact’

Rob Wiblin

‘We're currently making AIs that are getting smarter and smarter, and we don't know how to make them care about us’

Nate Soares

‘I'm just feeling this impulse in myself to not take this seriously … I think that's a defense mechanism’

Kevin Roose

‘We're only a couple years away … from AIs that are as good as the best humans at AI research’

Dwarkesh Patel

‘We're training them so we can decide what we want them to care about’

John Schulman

‘We've had this quiet whooshing sound as AGI slides past’

Trent McConaghy