3 ms·
Re; Section 3.1 This is a question on how human do we want AI's to act, which I think could just be set thru system prompts. Section 3.2 I think this is an
by wagwang 1y ago
Re;
Section 3.1
This is a question on how human do we want AI's to act, which I think could just be set thru system prompts.
Section 3.2
I think this is an argument saying that AI's are fundamentally missing certain sensory inputs so its information space is limited? Bad argument cuz you can always amend sensory information. The question could also be reframed as an experiment design problem instead of treating AI as an oracle. There's no reason an autonomous reasoning system can't do this.
Section 3.3
This is probably the worst argument yet. It's basically claiming that AI can't synthesize information?! Idk why the author keeps trying to simulate AI with his own words instead of just running the systems outright.