3 ms·
> we still consider AIs to be 'things' Do we, though? You and I do, sure. Most people here will, probably. But at least one counter-example was on display a fe
by schroeding 4y ago
> we still consider AIs to be 'things'
Do we, though? You and I do, sure. Most people here will, probably. But at least one counter-example was on display a few weeks ago, the guy from Google that told the press that their text completion engine was "alive" and "had agency".
From my friends I talked about this (which are not in IT), most believed him. YMMV, but I seriously believe a good chunk of the population thinks we already have thinking A(G)I. I don't think there is a "we" here, anymore. :/
- seydor 4y agoDo you think you 'll see an AI in jail anytime soon? If not, then it's nowhere nearly "a good chunk of the population"
- schroeding 4y agoDo we see pets in jail? No, but most people wouldn't say they are things (even though, in German law, they literally are), they are more or less intelligent beings with agency. I don't see the correlation, to be honest. A good chunk (but still a minority) of the people believing something doesn't automatically change the law, anyway, does it? :/
- seydor 4y agopets do get euthanized/ muzzled if they are found 'guilty' against humans because we do think they have agency . Sometimes we blame the owner for things they could have done, but there are cases where it's beyond their control. For the same reason we don't jail kids but sometimes they do get punishment. We do thing kids and pets have agency, but not full agency. At the current stage i don't think there is any AI that can be punished, or anyone that would credibly claim that an AI must be punished. Its maker will always be punished instead. Well true but i think that chunk is quite small. It's one thing to nonchalantly say "this is alive" and a very different thing when you have to deal with the consequences.
- schroeding 4y ago> It's one thing to nonchalantly say "this is alive" and a very different thing when you have to deal with the consequences. Yeah, fully agreed! Chatbots and "creative" ML systems are in the weird spot where they can't physically kill or hurt people, like e.g. a self-driving car, and perform tasks that "feel" like they need intelligence. It's also absolutely quite possible that the "chunk" is way smaller than I think, I'm just blindly extrapolating from my social bubble :D