As in - an emergent phenomenon made possible by the presence of all three?
Nothing we’ve seen in the realm of ai touches even a speck of the brain’s complexity and ongoing re-configuration in the presence of change and in the context of survival.
I’ve been reading George Lakoff’s work - turns out our ability to choose is based on an emotional component. Any rational decision is based on a preference, and all preferences are detected via the mechanism of emotion.
People with certain brain traumas resulting in damaged emotional centers become unable to make choices.
You'd also have to drill down into the definitions of body, nervous system, etc. Is a keyboard part of a nervous system? Is a computer tower a body? Why or why not? Would your criteria necessarily exclude the possibility of artificial sentience in the first place (although I'm not necessarily saying that's a flaw)?
Arguing that they are not sentient is not in conflict with materialism. And in fact it is supported by tons of material evidence: for example they don’t act sentient at all if you decline to provide a prompt.
The argument is 'what is sentience?'
>And in fact it is supported by tons of material evidence: for example they don’t act sentient at all if you decline to provide a prompt
This falls into the category of 'appearance of sentience'. Based on this logic all that would be required would be a perpetually self-prompting InstructGPT instance.
The comment at the top of this thread, to which you replied, states “Chatbots are not sentient.” Which is a narrower conversation.
I have not seen one reason why that confidence should exist. People seem to think we are special because we are made of meat, and we are “more complex”. But I still ask — what is so different about the human brain than an LLM? And further, what proof is there that we know what sentience truly is, and that LLMs don’t have it?
AIs made from artificial neural networks are actually not quite the materialistic thing. It's some spooky emergent property that doesn't depend on material substrates, you can run an AI on a Nvidia card for sure, but you can also run it in your head (theoretically, if you live forever). Or at least with the assistance of (a lot of) pen and paper. Where does the sentience live then, in your mind? That doesn't sound very materialistic to me :D
(Note: I think the GP is full of it, but I just thought I'd set the record straight about materialism here :P )
The question, then, is: what are the properties of the system that the term describes?
In this case, it looks kinda like: memory, self-awareness, and understanding. Memory isn't a question - it's possible for these things to remember (even if sometimes that memory is in the form of access to data on the internet). Self-awareness is trickier - but by most measurable schema, the machines seem to be doing as well as humans do in this regard (they can describe themselves and talk about their place in the world in context). Understanding is a topic of hot debate - but I think it tends to fall to the same problems as that of sentience generally; that is, we have this concept of what understanding is, and we know enough about how LLMs work to say that they're not the same, but if you start poking at how to measure it, we're back to square one vis a vis what's apparent might not be what is.