5 ms·
We can’t even prove other humans are conscience, right? We just assume it because it would be silly to assume we are somehow unique. I think it will not really
by bee_rider 4y ago
We can’t even prove other humans are conscience, right? We just assume it because it would be silly to assume we are somehow unique.
I think it will not really be a sharp line, unless we actually manage to find the mechanism behind consciousness and manage to re-implement it (seems unlikely!). Instead, an AI will eventually present an argument that it should be given Sapient Rights, and that will convince enough people that we’ll do it. It will be controversial at first, but eventually we’ll get use to it.
That seems like the real threshold. We’re fine to harming sentient and conscious creatures as long as they are sufficiently delicious or dangerous, after all.
- drewcoo 4y ago> We can’t even prove other humans are conscience, right? are conscious We can prove many of them don't have consciences. Then again, they can prove we don't either.
- dusted 4y ago> We can prove many of them don't have consciences. Some recovered coma-patients would like a word with you. Proving the absence of something like that is going to be pretty difficult (teapot, god) because what we're capable of is usually proving the existence of something.
- vcxy 4y agoI think drewcoo's point was that "having a conscience" and "having consciousness" are not the same thing. They were pointing out that the wrong word was used with a joke.
- achow 4y ago> AI will eventually present an argument that it should be given Sapient Rights On the other hand, today when we see some sign of consciousness with other living beings - smart chimpanzees, dolphins, ravens.., giving 'sapient rights' never come into discussion. 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'? Maybe how much ever smart or humane an AI is, it would never be equal to another (anthropomorphized) living being.
- freshbakedbread 4y agoI’m not sure that’s an equal comparison. These other beings that have been researched to have human like consciousness have a core difference from the latest/future AI: they can’t talk. Now/soon, AI will be able to argue with us for its own sapient rights. We humans have also become more and more accustomed to text only communication that we’re psychologically prepped to accept an AI as a human (or other anthropomorphismes living being) once it shows emotion, memory, and reason. Maybe not even reason.
- DennisP 4y agoBut those other beings are based on hardware very similar to our own, which we know supports consciousness. They're just not quite as smart. We don't actually know that consciousness is a computation, that computer hardware can support it, or even if it can, that the algorithms used in our AI can be conscious. It's possible that an AI would be a "philosophical zombie," exhibiting intelligent behavior without any conscious experience or qualia.
- zamadatix 4y agoThe important part seems to be that the thing can convince us it's conscious not whether it is actually conscious in a similar way we are. We don't know that anything is actually a computation but that hasn't stopped us from using computations in place of real systems.
- kelseyfrog 4y agoIs this implying that consciousness could be supernatural?
- DennisP 4y agoNot at all. Maybe consciousness is associated with particular physical properties, or a configuration of an electromagnetic field, or some quantum effect. Maybe the IIT guys are right and it depends on physical feedback loops; digital computers actually have little of this sort of feedback, so would have little consciousness regardless of the complexity of their programming. Or maybe it's computation. But we really have no idea. Any of these would be compatible with materialism, but we haven't made any real progress in even conceptualizing how qualia can emerge from any physical system. Of course that could just be because we haven't figured it out. Philosophers of mind look at other possibilities too though. One approach is to say each particle has its own fundamental consciousness, and somehow this aggregates in larger complex systems. But nobody's figured out how that might work either. Then there's Kastrup, who argues that the only truly skeptical approach is full-fledged idealism, because qualia are the one thing we directly experience. But even that doesn't imply that anything outside the bounds of physics could possibly occur, so it's not necessarily "supernatural" even if it's not materialism. Assuming that qualia somehow comes out of a computation, without any sort of explanation, is at least as much a magical leap as anything else.
- dusted 4y ago> AI will eventually present an argument that it should be given Sapient Rights The fictional character Data already did this in Star Trek: The Next Generation, and he is about as real as any AI out there today, and since they've all been trained on a body of text that is sure to include many instances of Datas dialogue, they're already able to predict such an argument (along with the many other facets of such arguments present in whatever science-fiction writing they've been exposed to)
- bee_rider 4y agoI think it would not convince many people. The workings of an LLM are sufficiently well understood that most people would see this the replication of these arguments for what it is; not an independent thought.
- thfuran 4y agoHave you met any people?
- bee_rider 4y agoHahaha, fair enough. Usually the logic part of our brains gets cut-off in the direction of maintaining the status quo, though, so I think an AI would actually have to come up super-compelling arguments if it actually wanted to have rights. I mean we become emotionally attached to everything, so an AI car, or whatever, will probably be treated like a pet. But getting human treatment will probably be a higher bar.
- Fatnino 4y agoI think the only restriction on eating something delicious should be if it's the same species as you. If humans discover/encounter some other species that happens to be conscious or sentient or even more intelligent than humans, that other species is fair game. We don't get moral outrage at a tiger for eating an ostensibly more intelligent human. It's just the way tigers are. We do try to get even though. Because that's just how humans are.
- bloppe 4y ago> unless we actually manage to find the mechanism behind consciousness and manage to re-implement it (seems unlikely!) Re-implementing it may be more likely than you think. The field of connectomics concerns itself with modeling natural brains and is currently constrained by scaling problems, much like genomics was a couple decades ago. As computing power continues to grow, it's entirely likely that humans will eventually be able to simulate an actual natural brain, even if that does little to further our understanding of how it works. The current state of the art in AI is attempting to reach consciousness via a different route altogether; by human design. Designed "brains" and evolved brains have a crucial difference; the survival instinct. Virtually all of ethics stems from the survival instinct. A perfectly simulated survival instinct would be ethically confusing to be sure, but the appearance of a survival instinct in current LLM's is illusory. An LLM plays no role in ensuring it's own existence the way us and our ancestors have for billions of years.