4 ms·
I've recently been enjoying thinking of GPT-3 utterances as creating short-lived consciousnesses for the length of an interaction. There's no consciousness duri
by sdenton4 4y ago
I've recently been enjoying thinking of GPT-3 utterances as creating short-lived consciousnesses for the length of an interaction. There's no consciousness during training (just filling in the blanks...) but when we interact with it in a generative fashion, there is - depending on your prompt - a somewhat coherent 'I' that is invoked, and maintained over the course of the conversation. Seeing as I have no strong claim on a strict definition of consciousness, Imma gonna go ahead and call it conscious.
the thing is, this is /commoditized/ consciousness, which can be spun up and discarded at will, a million times a minute. Totally incapable of coordination or planning, in contrast to, say, SkyNet. The future is usually weirder than we imagine.
- hoseja 4y agoI don't think it's maintained over the conversation. The recent history of the conversation is fed to the engine with each new interaction. So the potential flash of consciousness is even shorter than you think. Each reply is a new flash, without memory, only connected by the written content of recent text.
- sdenton4 4y agoAnd that's just fine! Quite reminiscent of 'Permutation City' in fact... (In which uploaded human consciousness is slowed down arbitrarily... And worse.)
- jb_s 4y agoIt's going to be even weirder once Google et al gives AI more and more agency to interact with people and objects the real world.