[visarga]

> Personally, I consider any GPT or GPT like technology unsuitable for any application in which truth is important . Full stop. The technology fundamentally, in its foundation, does not have any concept of truth

I think you got it all wrong. Not all GPT-3 tasks are "closed-book".

If you can fit in the context a piece of information, then GPT-3 will take it into consideration. That means you can do a search, get the documents into the prompt, and then ask your questions. It will reference the text and give you grounded answers. Of course you still need to vet the sources of information you use, if you give it false information into the context, it will give wrong answers.

[jerf]

I don't think you're right. Even if you add "correct" context, and in many of these cases "I can locate correct context" already means the GPT-tech isn't adding much, GPT still as absolutely no guard rails stopping it from confabulating. It might confabulate something else, but it still confabulating.

Fundamentally, GPT is a technology for building convincing confabulations, and we hope that if we keep pounding on it and making it bigger we can get those confabulations to converge on reality. I do not mean this as an insult, I mean it as a reasonable description of the underlying technology. This is, fundamentally, not a sane way to build most of the systems I see people trying to build with it. AI Dungeon is a good use because the whole point of AI Dungeon is to confabulate at scale. This works with the strengths of GPT-like tech (technically, "transformer-based tech" is probably a closer term but nobody knows what that is).

[mannykannot]

This hangs on what it means to "take it into consideration." If you gave me new information, I would attempt to see it in context, evaluate its relevance, and either update my positions accordingly or explain why I do not see it making a difference. If I saw difficulties doing this, I would ask for clarification, explaining what it was that was seemed difficult or unclear.

As far as I can tell, there is no reason to think that the way GPT-3 generates its responses could possibly result in this happening - even the basic ability of correctly inferring corollaries from a collection of facts seems beyond what those methods could deliver, except insofar as the syntax of their expression matches common patterns in the corpus of human language use. And the empirical results so far, while being impressive and thought-provoking in many ways, support this skepticism.