Jeffrey M Epstein

@jeffreymepstein.bsky.social

Quantum information, computation, and foundations. Increasingly interested in logic and the philosophies of physics, language, and mind. 📍 San Francisco jeffreymepstein.github.io/

There are two wolves. One always lies. One controls the lever of trolley track. They're in a state of superposition inside a box. On the trolley are infinite monkeys at infinite typewriters going to an infinite hotel. The trolley must first travel halfway, but before that a quarterway, and so on...

There's a lot in this thread & its multiple & various reply/quote offshoots that i disagree with, & am exhausted by, but i want to focus on 1 that i keep trying to press on: Let's stipulate LLMs *Can* do grad- & even expert-level math & we should prep for a revolution in the field… So what then?…

Zach Weinersmith@zachweinersmith.bsky.social · 4d ago

There's lots of stuff LLMs can't do, but if you want to disbelieve all the recent math, you have to believe a bunch of Fields Medalists with long track records are just straightforwardly lying about the thing they care most about. Here's a recent blog by Gowers: gowers.wordpress.com/2026/05/08/a...

Is anyone studying the "cognitive load" of reading LLM-generated text? The extra interpretative labor that presenting such text asks of us?

Disappointing to see people I agree with ideologically absolutely refuse to engage in good faith with people who also agree with them, but object to their making bad arguments in service of shared goals. Hard to win a rhetorical war with arguments your enemies can just dismiss as factually wrong.

What a splendid idea: two physicists with utterly different view of "reality" set up an adversarial collaboration where they attempt to understand one another's positions as well as they can, in an effort to promote fruitful exchange. More of this please! arxiv.org/abs/2607.16379

Realism about the external world: an adversarial collaboration

One of us is a realist and believes that reality fundamentally consists in an external physical world, governed by laws of physics. Observers are either emergent (physicalism) or exist in addition to ...

arxiv.org

I'm interested that I've never seen an Everettian response to that issue. (Saying "But notions of selfhood are already complicated by neuroscience" is not a response, but a rather desperate evasion.)

Exactly this But beyond that, too, I know he knows how this works, because he was an academic, too And the job doesn't involve embracing ideas that solidify "public trust"; it involves affirming ideas that are most supportable according to the methods and modes of inquiry in a given field

Aparna Nair@disabilitystor1.bsky.social · 2w ago

No, Tyler. Im angry because I know you know that the lack of public trust in US universities is the result of DECADES of media and political efforts--of which your whole grift is part. Quite the gall to be part of destroying public trust and then turn around and point to the lack of public trust.


Tyler Austin Harper
@Tyler_A_Harper
·
5h
What’s striking about the academics who are hysterically opposed to more ideological diversity in academia is that they have no plan for regaining public trust and don’t even seem to think they need it. Their only strategy is complaining that the media and right are mean to them.
Tyler Austin Harper
@Tyler_A_Harper
·
5h
I think increasing the representation of non-progressive ideas in academia is a necessary (though not sufficient) condition for regaining public trust. If you disagree, fine. But most of the people melting down have no alternative theory about how to regain public confidence.
Tyler Austin Harper
@Tyler_A_Harper
In fact, the bulk of angry responses seem to tacitly turn on the idea that public trust isn’t something academics should have to lower themselves to care about. The public owes the Ivory Tower trust and money, but that same public should also have no say in what goes on there.

"Why am I making the statue? The same reason why Michelangelo did the Pietà, or Michelangelo did the David, or Leonardo da Vinci painted the Mona Lisa. Why? Just to commemorate a moment, you know, a person." Among other problems here, I think those guys made sure they were gonna get paid *first*.

MisterJayEm@misterjayem.bsky.social · 2w ago

"Furnari is left with a 6ft 4in statue of a renowned racist, which he has spent $100,000 making, and which no one wants to buy."

Just finished Charles Taylor's "The Language Animal". A lot of interesting discussion about the (two-way) connection between language and thought, much of which I definitely haven't really understood. Need to now resist buying all the books he cites in footnotes that I put little post-its next to.

It's true that most resources are bad, but also importantly (and I actually am always saying this) when you read a bad piece of human written text you can usually assume that a reasoning process went into it, so there's something to be gained by muddling through. That's not the case for AI outputs.

Alom Shaha@alomshaha.bsky.social · 3w ago

My niche complaint of the day - seeing lot of teachers using AI to produce all sorts of "resources" without checking them. The AI is using shit resources as its input material because most teacher produced resources are shit to begin with.

I enjoyed The Odyssey (2026) but I have one historical nitpick. The phrase “Jiminy Christmas” is a minced oath for “Jesus Christ”. As the story is set around 1200 B.C. it is unlikely that Odysseus would have said the phrase even once, let alone the dozens of times he says it in the film

Don’t ask me why, but I came up with an ‘adversarial graph coloring game’, where players each color a vertex per turn, and the first player who is forced to color a vertex connected to a vertex they already coloured loses. Is this game and its complexity known?

Taking an intro pottery class now, so thinking about conic sections. Wondering whether I can view any sufficiently well behaved space of manifolds as a space of intersections of fixed higher dimensional manifolds (or ok orbifolds). What’s the relevant theorem here?