Yohan J John

Newspaper cutting: The owls are not what they seem ARLINGTON, North Military Road, 3000 block, 5:20 p.m. Nov. 10. Responding to a call about an injured owl on the side of the road, an officer found a large mushroom.

Morning Bluesky.

Quilicura, Chile, one of the communities I wrote about in EMPIRE OF AI, has launched a brilliant initiative to inspire more responsible AI prompting. Today, don't use AI; ask the townspeople instead: quili.ai. So heartened to see this creative act of resistance.

Thanks! I'll have a look.

... what prevents us from hallucinating whenever we think of high-level concepts?

This is a huge and fun area for speculation! If I could directly activate a "little people" category-detector that is (presumably) in some higher order cognitive-visual area, would it lead to the subjective percept? Perhaps through top-down activation?

And if that *can* happen...

Good point. I am aware of some of these. The specificity is what I find striking here: little people are a highly particular sort of hallucination that presumably requires quite a bit of "hijacking" of low-level visual circuitry.

'They saw them on their dishes when eating': The mushroom making people hallucinate dozens of tiny humans Only recently described by science, the mysterious mushrooms are found in different parts of the world, but they give people the same exact visions.

"With L. asiatica, though, "the perception of little people is very reliably and repeatedly reported", Domnauer says."

Wild. The implications for human perceptual systems are... hard to wrap one's head around.

www.bbc.com/future/artic...

Animals used to be the one I listened to the least often... But it now just bubbles up in my consciousness regularly. Those guitar solos on Dogs.

And the keyboards on Sheep!

Meddle!

A taste for whimsical dogs, beaches, and epic monomythic journeys?

This point about professional inertia is spot on, and applies to pretty much every institution.

Structure is necessarily more long-lived and stable than function. What's not necessary is the degree to which institutional structure leads to individual cognitive hysteresis (e.g., in science).

My prediction / hope for 2026: this will be the year we start seeing the theoretical neuroscience and #NeuroAI fields start embracing spiking as something beyond a poor man's ReLU. Spikes aren't just the brain's activation function but they are fundamentally different.
1/

🧠🧪🤖

the root of " #survive "
is "living-over"
and when I think about it
I have survived a good many things;
they have affected me,
yes,
they have torn me apart
the way ice tears potholes
from asphalt,
but I have lived over them
and laid fresh layers of life
over every crack,
no matter how wide.

#vss365

"However, as we lack a ground-truth understanding of how these complex models work, it can be hard to tell if our tools provide meaningful insights."

There's a disconcerting divergence between prediction accuracy and "explanation", broadly construed.

I am a great proponent of analogies in science, but I have gradually learned that the ability to jump from one analogy to another is perhaps more important than the ability to devise one in the first place: apprehending the latent invariant is the key phase transition, cognitively.

On the other hand, any conclusions we draw will have a very short shelf life, because the machines are in constant flux. Meanwhile, the ease with which one can do this research will sap the attention of researchers away from the harder work of understanding humans.

The belief that an LLM can help you write science reveals such magical thinking about language (and LLMs). If your writing is unclear, an LLM won't find its meaning hidden in the words. Only you can explain what you mean; the LLM does not read your mind. It completes with what *others* have written.

"A formalist is a man who can't understand a theory unless it is meaningless."

Some real gems in Saul Gorn's "Compendium of Rarely Used Cliches":

noemata.net/last+fawned/...

My mind has still not fully processed the fact that Werner Herzog has an Instagram account.

It's really something. Also check out Cameron Winter's solo album.

Heard the #Geese album after a break of a few weeks. It already sounds like an old classic.

As it turns out, it will also remove hyphens indiscriminately.

What rough beast, its hour come round at last, slouches...

I hope someone draws this wonderful story about Tom Stoppard's Arcadia and the power of the arts to help us see things differently, to the attention of our Education Secretary. Do read it, it will lift your spirits.

I completely agree: there really ought to be work on this, but I haven't come across anything.

I like some of Schwitzgebel's work, so I'll check it out! Thanks!

This from @davidbatherwoods.bsky.social's biography of Schopenhauer is sending me

YouTube video by Like Stories of Old Why Movies Just Don't Feel "Real" Anymore

Really like the concept of haptic visuality.

My favorite kinds of art criticism have much in common with phenomenology, broadly construed.

youtu.be/tvwPKBXEOKE?...

To what extent is ChatGPT like Indecisive Dave from the Fast Show?

I sometimes ask for book recommendations & when I question it, it says, “Yeah, you’re right - rubbish book.”

I've tried googling new papers but nothing seems to confront the strangeness of the present moment.

Have there been any major new theoretical papers in cog sci and/or philosophy of mind about human conceptual structure inspired* by large language models?

__
* the inspiration could well be 'adversarial'. 😛

#AI #philosophy #cogsci