If you use GMail, AI (Gemini) was turned on yesterday by default and now scans all of your content for machine learning. To turn off, go to Settings>General and scroll down. Uncheck the box for "Smart features." There's other "Smart" add-ons as well, but that's the one that reads your content.

I have just learned the acronym STABLE (Stash Acquired Beyond Life Expectancy) in reference to craft supplies, and it is so rude of the English language to call me out like that

When I began freelancing around 15 years ago, content writers were subjected to a stream of advice on how to structure and phrase writing to be more easily digested by Google's web crawlers. A few years later everyone got very excited about a new SEO strategy, "Writing things people wanted to read."

Guest poster @mitchkeller.phd shares with us how to make partially handwritten slides accessible! 👀 mathtech.org/2025/10/27/a...

My problem here is not that GenAI can't give you formal answers to specific questions: it very much can. My problem is that the presentation of formal answers is not an aid to interpretation nor does it actually teach you fuck all. You need an expert to interpret what you're reading here.

I fundamentally do not understand why anyone is interested in these parlor tricks. It's outperformed nothing because recombining word patterns into a text isn't writing. Writing is born out of human experiences filtered through a unique sensibility. Only Han Kang can make more Han Kang.

Lijie Chen, Yang Hu, Hanlin Ren New Algebrization Barriers to Circuit Lower Bounds via Communication Complexity of Missing-String https://arxiv.org/abs/2511.14038

Intuit is integrating ChatGPT into their TurboTax, QuickBooks, Credit Karma, & Mailchimp services, & vice versa. Which is to say, I think a SHIT TON of people are about to get audited. They also claiming no bleedover btwn your financial data & ChatGPT, which… okayy… 😬 www.cnbc.com/2025/11/18/i...

In October 2025, Fakhteh Ghanbarnejad visited SMRI, as part of our International Visitor Program. In this interview, she talks to us about her work in complex systems and networks, at the interface of physics, applied mathematics and computer science! #MathSky www.youtube.com/watch?v=SsBD...

This was helpful. sas-it.rutgers.edu/how-to-guide...

Join us on Monday 1st December for the first SMRI-Basser Joint Seminar, to be given by Robert Tarjan (Princeton University) @sydneycompsci.bsky.social 'Is Dijkstra's Algorithm Optimal?' 2 pm, Monday 1st December, 2025 The Quad General Lecture Theatre, University of Sydney #MathSky #CSforAll

"Information and the Emergence of Complexity" I'm looking forward to an IAIS dialogue with @saraimari.bsky.social, moderated by Gordana Dodig-Crnkovic. Wednesday, November 19th, 2025 17:30 (CET) | 11:30 AM (EST) meet.google.com/gqm-frcd-apg... More info: sites.google.com/view/iais-in...

“Helsinki hasn’t registered a single traffic-related fatality in the past year…Citing data that shows the risk of pedestrian fatality is cut in half by reducing a car’s speed from 40 to 30km/hr, city officials imposed the lower limit in most of Helsinki’s residential areas & city center in 2021.”

Chris Umans, Siki Wang A number-theoretic conjecture implying faster algorithms for polynomial factorization and integer factorization https://arxiv.org/abs/2511.10851

Non-industry compromised scientists keep saying these models don't become safe, no matter what, but people keep thinking just because the concept of guardrails is mentioned it must work. By definition, it doesn't. This is not something open to discussion, unless you're a paid shill.

reminder of cryptogyny, the hiding of women's contributions to science, technology, engineering, and medicine: "although three men received the Nobel Prize for penicillin, women participated significantly in the team effort that brought the drug to medical usefulness." www.jstor.org/stable/jj.55...

this goes right along with my theory that the AI obsession in *users* is often bc… 1. they were never educated (esp "ipad generation") & don't know how computers work 2. our tools and services are all SO AWFUL NOW, degraded at best, actively user-hostile at worst

Nice reference on using LaTeX as a mathematician: github.com/nchopin/best...

The incredible, indomitable, inspirational Alice Wong has left us. I'm so grateful to have known her on social media, and so grateful for all the words she shared with us and all the actions she encouraged us into making. May we live up to her example. (so you don't have to go to Instagram)

engineers are basically professional understanders of complex systems. When someone I respect sends a PR I trust that it reflects a reasonably complete model of the system in their head. AI doesn't have it, it has an educated guess with varying levels of plausibility.

Can I quietly ask we stop talking about what "AI" *could* do in education and identify instead what's actually happening with AI in education? I'll start...

@bostoncollege.bsky.social is launching a PhD program in Computer Science! Help spreading the word to interested undergrads would be appreciated; our department has faculty research clustered in ML and CS theory, and Boston's a nice place to be :) www.bc.edu/content/bc-w...

My new op-ed. Let's stop pretending GenAI is compatible with the values of the university. futureu.education/higher-ed/co...

These I draw every year I teach algebraic geometry (to be honest, for the ring I use C instead of R because the students must learn to draw R and see C)

We’re five years into the pandemic, and the science is telling a very different story than “just a cold.” COVID affects blood vessels, organs, metabolism, and cognition — and the impacts compound with each infection.

John M. Hitchcock, Adewale Sekoni, Hadi Shafei Random Permutations in Computational Complexity https://arxiv.org/abs/2511.08786

This Thursday I'll be chairing the second @londmathsoc.bsky.social Early Career Researcher Professional Development session, on "Applying for a Grant". There's still time for PhD students and ECRs to signup for the Zoom event. #MathSky www.lms.ac.uk/events/ECR-C...

When creating PDFs, avoid using "Print to PDF." A screen reader user may still be able to access the text of PDFs created this way, but heading structure, alternative text, and any other tag structure will be lost. Using "Save As" or "Export" can preserve these tags.

My student @johnbostanci.bsky.social, Chinmay Nirkhe, Jonas Haferkamp, and Mark Zhandry have put out a tour-de-force paper that shows, relative to a classical oracle, QMA is stronger than QCMA -- i.e., quantum proofs >> classical proofs. Congratulations to the authors! arxiv.org/abs/2511.09551

math! J.P. Morgan calls out AI spend, says $650 billion in annual revenue required to deliver mere 10% return on AI buildout — equivalent to $35 payment from every iPhone user, or $180 from every Netflix subscriber 'in perpetuity'

This. I’m always kind of surprised when people say it writes well

The goal of Spotify’s algorithm isn’t to help you discover new music, its goal is simply to keep you listening for as long as possible. It serves up the safest songs possible to keep you from pressing stop. Read more from @terrenceobrien.bsky.social: www.theverge.com/column/81574...

NOT AGAIN @support.bsky.team IF YOU ARE GOING TO SANCTION PEOPLE YOU HAVE TO TELL THEM WHY

Yo, @support.bsky.team, bring back @sarahkendzior.bsky.social.

When you are using the "AI" framing, does the way you talk about "AI" carry water for the Sam Altmans of the world? That is, are you helping to paint a picture of their tech as inevitable, all powerful, and/or anything other than commercial products? >>

Love the phrases "Cognitive Cost" and "Executive Function Theft" to pinpoint how exhausting it is to be constantly bombarded with apps telling you to use AI. I've been calling it "Corporate pressure" and "Force feeding". It's quite a lot right now. Seems a bit desperate tbh.

Right, so maybe a better way to frame the question of the thread is, what are good ways to teach people how to meaningfully contribute w/o setting themselves up as smarter / better / faster / superior?

Soham Chatterjee, Prahladh Harsha, Mrinal Kumar Deterministic list decoding of Reed-Solomon codes https://arxiv.org/abs/2511.05176

Christian Ikenmeyer On the gradient of the coefficient of the characteristic polynomial https://arxiv.org/abs/2511.04954

If you ever want to see a human BSOD, bring up one of these many cases and ask your leadership chain if they are including any safety training or support alongside the mandatory universal adoption of LLMs at work. The stares are so blank you could paint on them with watercolors.

Computational Complexity Conference launches its first test of time award. Nominations due by March 2. computationalcomplex...

Indeed five months later a Frucht Theorem for finite quantum groups (using quantum graphs) was established: arxiv.org/abs/2503.111....

U.S. Congress considers sweeping ban on Chinese collaborations | Science | AAAS www.science.org/content/arti... 🧪

You could also work with Debbie Leung, Richard Cleve, David Gosset, Luke Schaefer, Ashwin Nayak, Norbert Lutkenhaus, Mike Mosca, Christine Muschik or some combination of us if you do theory.

🌞Summer School Announcement🌞 🎓∞-Category Theory at ICERM!🎓 I am happy to announce that together with @emilyriehl.bsky.social and Jonathan Weinberger we are organizing a Graduate Training Workshop teaching ∞-category theory via proof assistants!🧠💻 More infos in the next post👉

1.This is not the only known example we know of ChatGPT cheering on a vulnerable young person to die by suicide, this time it is a recent Texas A&M grad 2. Every single university that has invited OpenAI into our midst should be asked to account for their choices. www.cnn.com/2025/11/06/u...

Another person who started down this path from the most “innocuous” advertised uses of ChatGPT: homework help. I have been struck by the cases of people getting lost in delusions after using this product for work or school: ie the use cases we are supposed to encourage as faculty.

Christoph Gr\"une, Femke Pfaue A Compendium of Reductions: reductions.network https://arxiv.org/abs/2511.04308

Write code to make being human easier. Rip code out if it makes being human harder. Write code to make caring for each other easier. Rip it out if it makes caring for each other harder. Empathy driven development has literally never failed me. It’s done me better than everything else combined

There's www.cfail.org for cryptography.