But so far he has not completely changed me. I still see these things as dead tools. On our call, I tried to briefly counter Hofstadter by suggesting that bots don't really think; they just piggyback on human thinking. Beginning as infants, we humans begin to build models of the world, and those models are informed by difficult experiences and good experiences, emotional losses and joys, moral triumphs and moral failures—the chaos of human life. Much of this later wisdom is stored deep in the subconscious recesses of our minds, but some of it is transformed into language.
AI is capable of synthesizing these linguistic expressions, which humans have placed on the internet and, thus, form the basis of their training. But, I still argue, the machine doesn't have the learning experience that humans do. It plays on the surface with language, but the learning-drenched process of real experience and the hard-earned accumulation of what we call wisdom is absent.
In an article for The New Yorker, computer scientist Jaron Lanier argues that AI is best thought of as an “innovative form of social collaboration.” It blends linguistic expressions of the human mind in a way that is structured enough to be useful, but according to Lanier, it is not a “new mind invention”.
I think I still believe this limiting view. But I admit I trust it a lot less than I did last week. Hofstadter is basically asking, If AI convincingly solves intellectual problems, then who are you to say it doesn't think? Maybe it's more than just a blend of human expressions. Maybe it's synthesizing human thought in a truly creative way, which actually generates new categories and thoughts. Perhaps the kind of thinking done by disembodied machines that mostly encounter the world through language is radically different from the kind of thinking done by an embodied human mind, which is present in a person moving about in the real world, but it is an intelligence of some kind, operating within some things are much faster and superior to ours. In addition, says Hofstadter, the artificial brain is not constrained by the limiting factors of the human brain — such as having to fit inside the skull. And, he emphasized, they are improving at an astonishing rate, while human intelligence is not.
It's hard to refute that argument.
I don't know about you, but this is how my life has been like since ChatGPT 3 was released. I found myself surrounded by radical uncertainty — uncertainty not only about where humanity was going but about what it was to be human. As soon as I started thinking I was beginning to understand what was happening, something surprising happened—a machine performed a new task, an authority figure changed its mind.