Ask HN: What would it take for an AI to convince us it is conscious?

Re:Ask HN: What would it take for an AI to convince us it is... by wruza on Sunday February 19, 2023 @07:30PM (#34863479)

How do we recognize that an AI has become much more 'conscious' than a very very smart chimpanzee, so that it should get the 'sapient right'?

The premise is off. We do that when it’s clear that it/they can take responsibilities that come with these rights. It’s not a blessing, it’s a contract. Chimps and dolphins couldn’t care less. Some individuals too, but we tolerate it because… reasons.

[ Parent | on HN ]

Re:Ask HN: What would it take for an AI to convince us it is... by VierScar on Friday February 24, 2023 @06:13PM (#34931573)

What do you mean we tolerate? If you mean criminal behaviour we don't really tolerate that. If you mean kids or disabled I think I've heard it justified that we have a contract to their parents that we still respect their rights.

For disabled I guess you could say it's a contract with the rest of society? Because we don't like the idea of treating other humans below a certain threshold. Or.. reasons I guess haha

[ Parent | on HN ]

Re:Ask HN: What would it take for an AI to convince us it is... by wruza on Friday February 24, 2023 @07:17PM (#34932201)

In a sense that all humans deserve a trial, relative safeties while at it and other things claimed as fundamental rights even if they tend to commit a crime as a lifestyle. A misbehaving animal or an AI that cannot be shown to understand the principles reliably (i.e. en masse of their species) would be simply “turned off” or get isolated with much less bureaucracy.

To make it clear, I’m not opposing these reasons, it was just a note.

[ Parent | on HN ]