Human Routers of Machine Words
When I open a link, say on Hacker News, and I see a blog post or a GitHub README<br>obviously written by AI, I feel a few things. I feel offended, because it’s like<br>I’ve been tricked, like the author thinks I’m a rube who won’t notice or mind. I<br>feel sad at how common this experience is, how many people are happy to dump<br>their sewage on the commons and sign their name on it. And I feel contempt for<br>the author, because if you use AI to write, you are a waste of biomass. Let’s<br>not mince words here. Someone who is so eager to replace themselves, that they<br>would have a machine write in their stead, when the machine can’t even write<br>good yet: what do you call that, if not contemptible? It’s like making yourself<br>into a eunuch so Claude can fuck your wife. I block these people on sight.
I see people defend this with: “the ideas are mine, the writing is the AI’s”. I<br>take this to mean they threw a bunch of incoherent bullet points at the AI for<br>it to denoise and render into paragraphs. There’s a few problems with this.
The immediate problem is, as we’ve established, the author’s an idiot. If you<br>are so stupid you can’t even turn some bullet points into prose, then your ideas<br>are probably worthless. I think that’s a sensible inference.
Then there’s this broader, I suppose philosophical problem, of the alleged<br>distinction between lofty “ideas” and mere “writing”, where writing is just a<br>tiny implementation detail. This is a very convenient distinction to draw,<br>because it’s unfalsifiable: if the AI’s output is slop, your “ideas” are still<br>good, it’s merely the writing that failed to convey them in their true<br>form (rather like people who say they’re smart but “don’t test well”: what use<br>is this secret intelligence?).
Now, where are these “ideas”? They are invisible, ghostly abstractions. I can’t<br>look inside your mind fortress and judge your ideas. The only thing that’s<br>empirically observable, that different agents can coordinate on and talk about,<br>is output: the writing.
But say this wasn’t true. Say we have something like a very high-resolution MRI<br>machine, and we know enough neurophysiology that we can interpret everything<br>about the brain, i.e., we can read mental representations from recordings<br>of nervous system activity and structure. These “mental representations”, do we<br>expect them to look anything like logic? Do we expect the brain to have this<br>firm, crystalline ontology, that ideas are sentences in some souped-up<br>first-order logic? Absurd. If we could look inside the brain, to see the ideas<br>“as they truly are”, we wouldn’t find beautiful hard-edged Platonic objects, we<br>would find a nebulous, contradictory mess of memory and feeling and<br>intuition. That’s what our ideas are: not logical sentences but dreams.
How do we refine these dreams into a useful form? Through writing. The process<br>of communicating your ideas to another mind forces you to concretize them, make<br>them precise, clarify your assumptions, more generally, it turns ideas from<br>vague ghosts to solid, physical objects that can be manipulated: here you<br>realize these ideas that seemed so solid are ill-posed or contradictory or<br>incomplete. These failures are necessary parts of thinking, because they teach<br>you two crucial skills: knowing which ideas to reject, and improving or<br>otherwise transforming ideas in search of better ones. By analogy to tree<br>search: you’re learning to discard bad nodes early, and to select which nodes to<br>go invest more search into.
Josef Weizenbaum has a great quote about this, in Computer Power and<br>Human Reason (p. 108):
[O]ften when we think we understand something and attempt to write about it,<br>our very act of composition reveals our lack of understanding even to<br>ourselves. Our pen writes the word “because” and suddenly stops. We thought we<br>understood the “why” of something, but discover that we don’t. We begin a<br>sentence with “obviously,” and then see that what we meant to write is not<br>obvious at all. Sometimes we connect two clauses with the word “therefore,”<br>only to then see that our chain of reasoning is defective.
I’ve experienced this with writing software many times. The reason ideas are<br>more attractive than their realization is that when some project is vague, airy,<br>ill-defined, you can imagine it has all the good traits, and none of the<br>bad. When you start concretizing, you realize that some of your ideas don’t make<br>sense, that some good traits are mutually exclusive, that some of your goals<br>impinge on the others. Anyone can imagine a programing language that is as fast<br>as C and as dynamic as Lisp, but when you sit down and think through what those<br>goals entail, you realize the design becomes contradictory. The goals pull in<br>different directions. You have to make trade-offs. You have to make decisions<br>which close off large volumes of design space, forever. The idea was a thousand<br>beautiful, contradictory things at once, but the concrete reality can only be<br>one thing.
The artifact...