Yeah, you're right. It seems like we both have a similar picture of what GPT-2 can and can't do, and are just using the word "understand" differently.
So would you say that GPT-2 has Comprehension of "recycling" but not Comprehension of "in favor of" and "against", because it doesn't show even the basic understand that the latter pair are opposites?
Something like that, yes. I would say that the concept "recycling" is correctly linked to "the environment" by an "improves" relation, and that it Comprehends "recycling" and "the environment" pretty well. But some texts say that the "improves" relation is positive, and some texts say it is negative ("doesn't really improve") and so GPT-2 holds both contradictory beliefs about the relation simultaneously. Unlike humans, it doesn't try to maintain consistency in what it expresses, and doesn't express uncertainty properly. So we see what looks like waffling between contradictory strongly held opinions in the same sentence or paragraph.
As for whether the vocabulary is appropriate for discussing such an inhuman contraption or whether it is too misleading to use, especially when talking to non-experts, I don't really know. I'm trying to go beyond descriptions of GPT-2 "doesn't understand what it is saying" and "understands what it is saying" to a more nuanced picture of what capabilities and internal conceptual structures are actually present and absent.
One way we might choose to draw these distinctions is using the technical vocabulary that teachers have developed. Reasoning about something is more than mere Comprehension: it would be called Application, Analysis or Synthesis, depending on how the reasoning is used.
GPT-2 actually can do a little bit of deductive reasoning, but it is not very good at it.
I don't think I am attacking a straw man: You don't believe GPT-2 can abstract reading into concepts, and I was trying to convince you that it can. I agree that current versions can't communicate ideas too complex to be expressed in a single paragraph. I think it can form original concepts, in the sense that 3-year old children can form original concepts. They're not very insightful or complex concepts, and they are formed by remixing, but they are concepts.
My thinking was that since everything it knows is something that was expressed in words, and qualia are thought to not be expressed fully in words, then qualia aren't part of what it knows. However, I know I'm on shaky ground whenever I talk about qualia. I agree that one can't be sure it doesn't have qualia, but it seems to me more like a method for tricking people into thinking it has qualia than something that actually does.
I liked how you put this. I've just posted my (approving) response to this on Less Wrong under the title "Does GPT-2 Understand Anything?"
There wasn't a large "manufacturing" sector for agriculture workers to move into, it became a large sector as the workers moved into it. Perhaps some current small sector of the economy will become a large sector as workers move into it? At least in the U.S., there's little evidence to support your claims of it being faster and more widespread-- jobless rates are at historic lows. Unless you mean it hasn't yet begun.
All that said, though, it is certainly the case that if you have a robot that can do anything a person can do, you don't need to hire any more people, and there must be some kind of curve leading up to that as robots become more capable.