Ignacio

This song was written in 20 minutes of pure rage after hearing how a senator got his son out of the draft for Vietnam...

Charlie Bennett and Gilles Brassard win the Turing Award for Quantum Information Science. Most notably they developed the theory for a provably secure quantum key distribution protocol. awards.acm.org/about/2025-tur…

New on "Proofs and Intuitions": Verifying Move Borrow Checker in Lean: an Experiment in AI-Assisted PL Metatheory. proofsandintuitions.net/2026/03/18/mov… The gist: I formalised Move's type system in Lean: 39KLOC, under a month, with Claude. Person-years in PL research are now person-weeks.

@rohanpaul_ai Who are the authors? Do they work for anthropic?

Anthropic's own study proves Vibe-Coding and AI coding assistants harm skill building. "AI use impairs conceptual understanding, code reading, and debugging abilities, without delivering significant efficiency gains on average" Developers learning 1 new Python library scored 17% lower on tests when using AI. Delegating code generation to AI stops you from actually understanding the software. Using AI did not make the programmers statistically faster at completing tasks. Participants wasted time writing prompts instead of actually coding. Scores crashed below 40% when developers let AI write everything. Developers who only asked AI for simple concepts scored above 65%. Managers should not pressure engineers to use AI for endless productivity. Forcing top speed means workers lose the ability to debug systems later. ---- Paper Link – arxiv. org/abs/2601.20245 Paper Title: "How AI Impacts Skill Formation"

My good friend Christian Coester solved a 50 years old open problem in self-organizing lists (was it really open? did people even care? yes and yes), and, you guessed it, ChatGPT-Pro made the key step in the proof*! (*a Christian Coester on the other side of the screen was needed to see through the full proof!) The problem is extremely simple: you are organizing (online) a list of items, and when item in position i is requested you pay i. Requests arrive i.i.d., and so the best thing to do is to organize the list in order of decreasing probability of appearance of items. But now what happens if the probabilities are not known, and you want a memoryless algorithm (note that with memory one could just estimate frequencies and approximate the optimal ordering)? 40 years ago it was shown that the most stupid memoryless rule of moving the requested item to the front achieves a competitive ratio of pi/2 (and this is tight). But empirically there seems to be a better idea, which is to be a bit less aggressive and simply move up the requested item by one position (the "Transposition Rule"), see Codex's experiment below. In the very first paper on this topic 50 years ago, Rivest conjectured that indeed this should be a very good rule. What did Christian prove? He showed that Transposition Rule's average cost is at the most the optimal fixed list plus ONE. Just a regret of one. Beautiful. In fact the proof shows something more delicate, and this is where ChatGPT comes in: Denote p_i for the probability of the i^th item (in the optimal ordering). Now under the stationary distribution of the Transposition Rule, denote s_i for the "excess cost" that item i incurs from wrong placement, namely: s_i = sum_{j < i} (p_j - p_i)*Prob(i appears before j) The regret of Transposition Rule is exactly the sum of those slack terms s_i. Now here is the magic: ChatGPT suggested that s_i might be smaller than p_i. If true that concludes the proof, since then we have sum s_i <= 1. Note that it is quite magical, I don't see a reason a priori why s_i would be smaller than p_i ... ChatGPT proposed a an argument to prove this for small n, and Christian found an actual proof for all n.

Short and wonderful read

Tristan Stérin used LLMs to hunt for bugs and inconsistencies in Rocq and Lean. This is actually pretty neat and kind of wild. twp.ai/4ixLIF

So it begins

Math, Inc. is proud to announce an all-star group of Veritas Fellows: Renowned professor Kevin Buzzard, alongside Fields Medalists Maryna Viazovska and Terence Tao. They will lead teams to build formal mathematics at unprecedented scale. 🧵

Acabo de mapear Ctrl a Caps Lock, no puedo creer la cantidad de años que viví sin esto.

@arielsbdar Claude haceme un port de Guile scheme para windows, no mistakes

La I.a. lleva a 0 el costo de software y del conocimiento. Con pocos dólares cualquiera sin años de experiencia puede crear y desarrollar cualquier cosa. Esto es uno de los shocks de productividad más grandes desde la revolución industrial. Muchos van a patalear e intentar retener sus quioscos, tapar el sol con los deditos, no van a poder. Solo van a sobrevivir los más aptos. No te duermas.

?

@JamesLucasIT Marais Les voix humaines m.youtube.com/watch?v=zCcuj9…

What's the most profoundly beautiful piece of music you have ever listened to?

@obc68 Al ángulo

Darío, con el mayor de los respetos y desde la experiencia en el sector aeronáutico, la teoría económica del costo marginal no se aplica de forma lineal en el aire. Combustible por Peso: En aviación, cada kilo cuenta. Un pasajero adicional (más su equipaje) altera el peso de despegue (MTOW). Esto aumenta el consumo de combustible por hora de vuelo. Multiplicado por las horas de una gira presidencial, el costo es real y medible, no es cero. Tasas y Seguros: Existen tasas aeroportuarias de aterrizaje y servicios de rampa que se calculan por pasajero o peso. Además, las pólizas de seguro tienen primas vinculadas a la cantidad de personas a bordo y su rol. Costo de Oportunidad y Logística: Un avión oficial no es un transporte privado personal; es un activo del Estado. El uso de recursos públicos para fines no oficiales (como trasladar familiares) rompe la trazabilidad del gasto y el protocolo operativo. Como empresarios, sabemos que lo que no se mide, no se gestiona,Darío. El "asiento vacío"en un vuelo comercial es una pérdida, pero un "asiento ocupado"por alguien ajeno a la operación en un vuelo oficial es un costo fiscal y ético innecesario. Abrazo

@xHaustedOfStats @littmath @davidmbudden Yes, I figured it was something like that, but which ones?

@_letrec @littmath @davidmbudden Budden was claiming they proved like 3 millenniam problems lmfao

I've just made a donation of $25000 CAD to the Against Malaria Foundation (through the Rethink Charity Foundation), the bulk of what I received from my bet with @davidmbudden. Thanks to David for the fun back and forth!

@cosmeluichito El problema es que vos pensas que los chicos sufren si no les das caramelos, salchichas y papas. Les pones milanesa cortadita, tomate cherry, queso, frutas y está todo bien, lo que les interesa es jugar, no comer como Baradel.

Lo que no entiendo de las prohibiciones de comida en los cumpleaños infantiles es que los cumpleaños no constituyen un hábito, son una excepción. Yo como todos los días claras, pollo y ensalada, pero si voy a un cumple como como Baradel. Dejá que se coma un tatín la creaturita.

Terry Tao, on the rate of latent errors in the literature: "I don't know, I would be very interested to see what the error rate actually is "Maybe we'll be pleasantly surprised, maybe unpleasantly surprised!

@oc_revolver @gptcrosa Que lindo trabajo

@gptcrosa Oportunidad de oro el que limpie el Slop, o que haga un agente antiagentes

Un gran problema con AI es que mucha gente puede automatizar el performative work e inundar de slop por todos lados