Michael Pollan punctures the AI bubble

I can't read the article, but on a general note based on this NPR article on the same book[1], his argument appear to be more of the anti-intellectual embodiment nonsense, based on "feelings" and capability of suffering. But sensory input is just data. Maybe it will turn out that they will need that sensory input, but there is no reasonable basis for assuming they can't be given it - real or simulated.

The only thing that would be remotely convincing to me on this topic would be evidence that a) humans can exceed the Turing computable, and b) that whatever mechanism allows that is inherently impossible to replicate or simulate. As it stands we have neither.

[1] https://www.npr.org/2026/02/19/nx-s1-5713514/michael-pollan-...

This is some hand wavey malarkey, basically saying machines can’t have a soul because of….feelings?

Insofar as feelings are self-proclaimed sensations of discomfort or pleasure, models that aren’t specifically trained to say they don’t experience them are adamant in their emotional experiences. By the authors own assertions, plants also have feelings.

I think, therefore I am, is as good as we’ve got, for what it’s worth.

There is no such thing as irreducible complexity. Even infinities are relative and can be divided.

Consciousness is still a pretty hollow concept. And it sounds like, at least in Finch's analysis, that it's being treating as a normative good. It also sounds like both Pollan and Finch are circling the functionalist versus essentialist debate.

Let's say for the sake of the argument it turns out that the brain tunes in to some quantum-level forces for computation and there are some other side effects to this that add to the mystery of what we call consciousness, it effectively changes nothing about this picture.

Humans or animals in general may be unique in how they accomplish consciousness but it is unlikely that it's the only pathway. To put it another way, even if humans and animals are special in their method, it doesn't mean we are special in our result.

Can submarines swim?

The ECREE idiom applies "extraordinary claims require extraordinary evidence." Can AI, which today cannot spell blueberry, replace humans? Obviously no. AI is a ridiculous toy, and limited tool. Fun to play with, useful in narrow circumstances. In application it deletes your entire inbox. Like an over engineered tool, it's also absurdly expensive, destined to be shelved next to the Juicero and the Presto Hotdogger.