Post

This is just nonsense, the New Yorker helping Anthropic out with some marketing. An LLM cannot plan, it doesn’t think, it’s not reasoning, it generates a sentence in a moment.

Have you used opus 4.6 in an agent harness yet? If not, then your opinion might be out of date.

Yes

You REALLY think you have a leg to stand on??

Yes. Why not?

bsky.app/profile/eryk...

Eryk Salvaggio@eryk.bsky.social

Functionalist approaches measure intelligence by how closely its outputs resemble the human mind. If you constantly recalibrate its outputs to resemble a human mind, you're not creating a thinking machine. You are creating a machine to trick other functionalists.

If it quacks like a duck and walks like a duck, is it for all intents and purposes, a duck? If by a hardline definition, one says no, then a plane does not fly if it doesn’t flap its wings? Philosophy aside, the latest models with agent harnesses are astonishing good at writing software.

Not just software for side projects, but production software. You have to know how to guide it, but it’s only possible in the last ~3 months. How you might categorize what it’s doing is almost irrelevant, as it’s automating a swath of what I do.

11/2/26, 8:44 pm0 reposts0 likes

The OP may not had her priors updated, due to not being a witness to the latest capabilities. My question was earnest.