Josh Specht πŸ‚πŸ  (@joshspecht) header
Josh Specht πŸ‚πŸ  (@joshspecht) profile picture

University of Notre Dame. Land and politics. Cattle and capitalism too. Hopefully another thing soon. Views are my own; retweets are not always endorsements

South Bend, IN
Joined January 2009
Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)Photo by Josh Specht πŸ‚πŸ  (@joshspecht)
Publication day for my new book, Property Values! Get it, read it, buy it for friends, sleep with it under your pillow, etc. press.princeton.edu/books/ha…
3
38
3
221
12,662
I can't explain it exactly, but the way this proof is being discussed here is kind of how people started reacting to surprising engine evals when machines took over chess. Humans have a logic to evaluating chess positions and engines could throw it all out and somehow be correct
Like most mathematicians, I believed the Riemann Hypothesis to be completely out of reach at present (even with the best AI tools), because I expected it to require substantial theory-building, probably in arithmetic geometry. The kind of new theory-building that fills entire books, and that AI isn't so good at yet. Mathematicians have a notion of how deep a problem is. A sufficiently deep problem is generally intractable until enough of the theory is in place. If difficult problems are like valuable mineral reserves buried miles underground, you first need to build the infrastructure to reach them. What Levent's proof seems to have done instead is something completely different. It takes an incredibly complicated tour through the infrastructure built for a hundred other mineral reserves, situated far from the destination, and somehow create a wormhole to the original destination. This is not theory-building in the usual sense. It is something much weirder and more alien. The best human mathematicians know how to build theory, painstakingly, to reach a goal. But no human brain, on its own, is capable of creating a wormhole that requires a 25,000 page long tour through existing infrastructure with no obvious or realistic connection to the goal. But huge credit to Levent here: I think no current AI, on its own, is capable of this either. I had never worked on or thought too much about the Riemann Hypothesis: it seemed very far removed from my own area of number theory. But the astonishing thing about Levent's claimed proof of RH (which I think has a very good chance of being correct) is that it uses a huge number of precisely the things that are central to my own corner of number theory: - Theta functions and other automorphic forms on unitary groups - Gan–Gross–Prasad period formulas - Relative trace formulas Etc., etc. A problem that I had always assumed would require new theories in arithmetic geometry may instead turn out to be tractable through an inhuman chain of existing (deep) results in higher-rank automorphic forms (and a lot else).
1
6
329
They couldn't do this in every type of position but as they got better and better they could do it in more and more human inexplicable cases
1
1
103
Chess could be a way of anticipating a world in which we are less and less relevant to knowledge creation. People still love chess at all levels even though engines are vastly superior to everyone. It hasn't killed the game. But everyone old enough knows some of the magic is gone
1
91
Josh Specht πŸ‚πŸ  retweeted
The Riemann Hypothesis has fallen. RH was the singular math result that had been named, in advance, as the obvious coal mine canary for dangerously advanced AI, over the last three decades. The methods used should generalize to "make AI very good at anything you can definitely test", and unfortunately that category includes "make AI better at making AI". This is like a private company announcing they have produced a million tons of 15% enriched uranium. It's not 90% enriched weapons-grade uranium, so you could proclaim the company's output is not immediately deadly right now. The concern is that the process line may generalize, to an extinction-level threat, within the unpredictably immediate future. Forecasters still had Riemann at 2035 one week earlier. This is distinct from earlier events in that earlier you could reasonably ask if AI was selectively picking off famous math problems that turned out to be easier for AI. The Riemann Hypothesis was the single challenge most often named in advance as a warning sign.
256
597
172
5,754
715,429
Josh Specht πŸ‚πŸ  retweeted
a lot of people are asking me if the noise you can hear in the background of this data center story is real. I guess I was skeptical too, before I went there in person. Yes. It really is that loud. It's like a big plane about to pass overhead, but it never does, which adds a weird feeling of anticipation and frustration. There's a low buzz, but also a higher-pitched whine that reminds me of an old fan. There's also a lot of construction noise, but of course that's varied and temporary. The construction is 24/7, and the site is illuminated at night by very bright lights that shine into some people's yards. The question I kept coming back to was: are people expected to just live like this? Forever?
338
3,061
273
12,158
1,055,804
Looking for my phone while actively scrolling on it. As timeless as wondering why my vision is blurry in the shower and realizing I'm still wearing my glasses.
8
290
Josh Specht πŸ‚πŸ  retweeted
I strongly believe that anyone who bans students from using AI should also refrain from using AI tools to grade or prepare their instructional materials. Either the human element matters, or it doesn't.
This should go over well: Berkeley is piloting an AI grading tool that grades hand-written work. It is β€œintended to decrease grading time…by 40-60% as written exams become increasingly common among professors attempting to fight back against the rise of AI.” An MBA student created the app after being assigned as the sole teaching assistant for a class with 133 students. It β€œwill be piloted this spring semester with up to 15 faculty members at the Haas School of Business.” β€œCourse instructors must review Rubrica’s work before its grades are released.”
15
9
3
106
5,225
The best and wurst of twitter is that someone always has a funny topical joke and it's better than yours.
1
1
179
I absolutely crush it in the meatspace but who can keep up with this???
1
1
98
What are y'all doing all day or did you prewrite these GTA VI jokes like a famous person's obit??
1
75
What if Groundhog Day was literary fiction? Read it!!
157
Just got my copy in the mail. This series is really really really good. Short books too, so easier than my exhortations to read Life and Fate. I've already preordered the translation of the next and final volume!
1
147
Ahead of Anthropic’s IPO, a rare coalition across AI safety, open source, e/acc, science, and law has signed the MisAnthropic Statement. Read the statement or sign at: misanthropicstatement.com/
1
203
Josh Specht πŸ‚πŸ  retweeted
The A.I. conscientious objector movement is growing, especially on campuses. My new story on how colleges are responding, and what this means for students/faculty who are refusing to use an ever-more pervasive and powerful technology. nytimes.com/2026/11/18/us/an…
29
232
37
931
124,537
when they say this, believe them
I still find it totally wild that the chief scientist of OpenAI posted this on the company blog, and we all just moved on to the next day's news
1
216
well, thanks to these bonkers enthusiastic reviews i lasted one day with my whole "i'm an adult who doesn't have time for GTA VI." Now I'll be "an adult who doesn't have time for work."
1
10
358
wait maybe i only lasted 0 days but who's counting
1
44
You think GTA VI is fun? Try Property Values: the Rise and Fall of the American Homeowner by Joshua Specht #soreal #realtalk #realfun
4
232