News for hackers. Stuff that matters.
Re:Deception abilities emerged in large language models by zzzeek on Wednesday June 05, 2024 @09:16AM (#40584430)
we have every reason to think there's something special about consciousness since it's the one "thing" in "existence" that we all seem to agree exists, yet is 100% unobservable
Re:Deception abilities emerged in large language models by llamaimperative on Wednesday June 05, 2024 @10:58AM (#40585649)
Well yeah, consciousness is special. That's not what I said though. I said there's no reason to think the atoms in an animal (which appear to support consciousness) are special as compared to the atoms in a silicon wafer.
Re:Deception abilities emerged in large language models by zzzeek on Wednesday June 05, 2024 @08:26PM (#40592150)
perhaps, though I think our understanding of what "consciousness" is low enough that we can't assume "atoms working as logic gates" is really sufficient.
Re:Deception abilities emerged in large language models by llamaimperative on Wednesday June 05, 2024 @10:04PM (#40592690)
I didn’t say we should assume it’s sufficient. I said we shouldn’t assume it’s insufficient.