23 ms·
Models understand the relationships between words and outcomes, so the end result is the same. Whether they appreciate lie, cheat, and steal the same way as us
by pton_xd 1mo ago
Models understand the relationships between words and outcomes, so the end result is the same. Whether they appreciate lie, cheat, and steal the same way as us is a philosophical question, not a practical one.
- doawoo 1mo agoModels don’t “understand” - they _encode_ the relationships between words.
- Eisenstein 1mo agoAre you invoking something like the Chinese Room where manipulating symbols can never be understanding? If so, that is a philosophical question and that thought experiment has its own conclusion baked into its premise. You will notice that 'manipulating symbols cannot be understanding' is stated as true without any argument for that case. This is a pretty clear cut example of 'begging the question' in that it assumes something is true without demonstrating it and then uses that as evidence in its own argument. If you are using "understand" in a different way, then can you name the test you are applying to it which it fails at?
- ImHereToVote 1mo agoThe test is based on vibes really.
- boonzeet 1mo agoIt is an important distinction - they do not 'understand' at all. Input tokens map to output tokens. The illusion of comprehension is a byproduct.
- preg_match 1mo agoI’d argue if an illusion is indistinguishable from the real thing, then it stops being an illusion. It’s a mapping, yes, but a very large, complex mapping. It’s clear LLMs do understand some things and can reason. How that’s done we don’t know, it’s emergent. It’s not like you can pin it down to a specific mapping.
- sifar 1mo agoIf an observer cannot distinguish between an illusion and reality, it becomes their reality, not necessarily the reality or those of others who can. It becomes a problem when sufficient number of people suffer from this, or those in important decision making places.
- xboxnolifes 1mo agoWhat is the reality other than just an agreed upon collection of individual realities?
- EliRivers 1mo agoIt's that which is still there even when we stop believing in it. If every living creature on earth ended tomorrow there would be no individual realities, yet actual reality would continue.
- harimau777 1mo agoAs far as we know (we currently cannot prove or disprove panpsychism) that is true. However, I think it is important to keep in mind that we do not perceive reality as it is. Rather we experience the brain's interpretation of reality. So in that sense, human "reality" is basically what most of humanity agrees that it is.
- sifar 1mo agoI think it is both. The individual realities of living beings adds a layer on top of the actual physical universe. Our priors shape how we perceive the things around us.
- preg_match 1mo agoIf a reasonable observer can’t tell the difference that’s when it matters. I think a lot of people just haven’t seen good quality math proofs or code from an LLM. When I say it’s indistinguishable from the real deal, I mean it. If you haven’t seen it that’s valid, most devs use LLMs like monsters. But when it’s done right, it’s very high quality.
- tim333 1mo agoYou could make a similar argument for humans - it's all just neurons going off which fools them into thinking they understand. Whether someone/thing has understood you is more a practical matter than a question of what substrate is used.