Hallucinations of "People From Humanity" After Communicating With "Artificial Intelligence"

There are meticulous, tenacious people who have learned to squeeze genuinely serious answers out of smart chats. They're in no hurry to share their method — for each of them, it's a personal competitive advantage, a source of professional authority. And there's a huge mass of users who mostly mess around with AI doing nonsense: asking it to do their work for them, trying to needle it, asking primitive questions without supplying important context — and getting predictable nonsense back, because the system doesn't know what's critically important for its answer. The goal of this article isn't to pass judgment on either of these groups, but to show: as long as we keep talking about "AI" as a single phenomenon, we're comparing things that can't be compared.