Andrés Villaveces Niño

Here, by French mathematician David Bessis, an extremely thought-provoking essay on the current deep change in what he calls the "theorem economy". Bessis takes issue with some claims by Geoffrey Hinton and also partially with some by Terry Tao. Beyond that, he offers some possible ways of dealing with the issue (he also takes a very critical look at what he terms "Hardy's curse", the brutal centrality of theorem proving in mathematics).

An extraordinary writing.

https://davidbessis.substack.com/p/the-fall-of-the-theorem-economy

Michael Rabin (Israeli mathematician, computer scientist, and recipient with Dana Scott of the 1976 ACM Turing Award for their work on computational complexity, also Saharon Shelah's doctoral advisor at the Hebrew University of Jerusalem, and later professor at Harvard University) has passed away. Honored be his memory.

Good news: my book on forcing, in Spanish (A través del universo - Una guía a los axiomas conjuntistas, consistencia, forcing e independencia) has appeared!

The book (a result of teaching several times forcing in various places) was beautifully published by Editorial Universidad Nacional de Colombia (Facultad de Ciencias) and Editorial Universidad El Bosque.

Mathematicians: please consider signing this petition to move the ICM 2026 out of the United States.

https://docs.google.com/forms/d/e/1FAIpQLSdHJhc8X83b8oL6rH2KDX0I730eraum5I8_IlWY23F82mHuag/viewform?

Con mucho orgullo, con mucha felicidad, con mucha participación de varias generaciones, celebramos nuestros 70 años del Departamento de Matemáticas, Universidad Nacional de Colombia el viernes pasado.

¡Mil gracias a todos/as quienes hicieron posible esta construcción de siete décadas, y esta celebración tan bella! Las fotos capturan la auténtica y profunda felicidad de ese día, de tantas edades distintas compartiendo un momento importante.

Celebramos el Día Mundial de la Lógica 2026 en Bogotá el 16 de enero la próxima semana/
We celebrate World Logic Day 2026 in Bogotá on 16 January, next week:

​Lectures by / conferencias de

Juliette Kennedy (University of Helsinki)
Jouko Väänänen (University of Helsinki)
Fernando Zalamea (Universidad Nacional de Colombia)

​See info here:

https://sites.google.com/view/bogotalogica/eventos/logic-day-bogot%C3%A1?authuser=0

Asaf Karagila gives a very basic (and very effective; as Santi Spadaro says, Asaf should be on prime-time TV) account on ordinal numbers.

https://www.youtube.com/watch?v=47qEMTMKRdA

OpenAI acaba de anunciar que uno de sus modelos logró refutar la conjetura de las distancias unitarias de Erdős.
Los matemáticos que revisaron la demostración señalan que no es un acierto por mera fuerza bruta, se trata de una generalización bastante ingeniosa de la construcción original de Erdős. Al parecer, la ventaja clave de la IA radicó en su poder de correlación para conectar de forma natural herramientas de áreas distintas (en este caso, la geometría discreta y la teoría algebraica de números). Señalan que es previsible que la IA, al explotar estas correlaciones entre áreas distintas de su enorme base de datos (casi toda la matemática publicada), empiece a obtener resultados novedosos conectando campos lejanos. (Lo anterior, escrito por John Jaime Rodríguez, colega mío en la UNAL.)

Agrego: Un logro impresionante para una IA. El artículo parece [hasta donde veo] invocar el "modelo interno" de OpenAI para hacer una construcción de un campo que termina dando el contraejemplo. La historia tiene gente muy ilustre: Erdős lanza la conjetura en 1946, con una cota específica demostrada, que luego mejoran personas como Spencer, Szemeredy entre otros. En 2010, Guth y Katz mejoran de manera impresionante la cota. Pero ahora la conjetura original de Erdős se cae, con la construcción ayudada de manera esencial por la IA.

Aquí un enlace al blog de combinatoria de Gil Kalai, donde se discute lo anterior:

https://gilkalai.wordpress.com/2026/05/21/amazing-erdos-unit-distance-problem-was-disproved-it-was-achieved-by-ai

With María Clara Cortés, we will participate this coming Thursday and Friday in the event Pedestrian Space, at the Boston University Center for the Humanities.

Our lecture will be given from Bogotá (we are both invited in person to Boston University, but we decided a few months ago to decline being physically present in the United States for this conference, considering especially the horrific situation with crossing borders at this time in history). The organizers gracefully accepted our request to give the lecture from here. We will miss the fantastic opportunity to be in Boston, a city we both like very much, and being fully invited for a few days. But we believe this is a moment in history when such decisions may be important, at least for us.

We will give ONE lecture together, called

Transits on Pedestrian Space: from factual to critical; a conversation between a mathematician and an artist.

Our lecture will be on Thursday morning.

https://www.bu.edu/humanities/pedestrian-space/

📐 DiscreMath invita

Este miércoles 25 de febrero (10:00 a.m.–1:00 p.m.), jornada de charlas cortas con László Szalay y László Németh, en el marco del convenio UNAL–Universidad de Sopron.

📍 Auditorio Juan Herkrath Müller
🔗 Inscripción: seguir QR

Un curso de extensión en la Universidad Nacional de Colombia (Departamento de Matemáticas)

A cargo de Juan Daniel López Castaño - estudiante de doctorado en Johns Hopkins, egresado de Matemáticas y Ciencias de la Computación de la Universidad Nacional de Colombia.

Pensamiento geométrico para machine learning y ciencias de datos.

Fortalece tus bases matemáticas y comprende cómo la geometría juega un papel clave en el análisis de datos y los algoritmos de machine learning.
📅 Del 9 al 13 de marzo de 2026
🧑‍🏫Modalidad: híbrido
🎯 Dirigido a: estudiantes, profesionales e interesados en ciencia de datos y machine learning.

https://ciencias.bogota.unal.edu.co/educacion_continua/cursos_diplomados_eventos/curso-de-pensamiento-geometrico-para-machine-learning-y-ciencias-de-datos/

With María Clara Cortés, we will participate in this event (Pedestrian Space) at the Center for the Humanities, Boston University: https://www.bu.edu/humanities/pedestrian-space/

The event is connected to The MA Collective (MA for Mathematics and Arts, but also as a tribute to MA (an important maximality principle) in set theory).

We will not travel to Boston this time (although we were invited): we will speak remotely, from Bogotá. We decided this several months ago, when travel to the USA became a hazardous issue.

Our lectures for Pedestrian Space will address the problem raised by Fred Sandback's thought and art works: the issue of the space of sculpture made _pedestrian_ by allowing the viewer to enter the sculpture, to become part of it.

We will take issue with this problem, from art and mathematics perspectives, and the inherent politics of pedestrian space.

There will also be a screening of two artworks by María Clara Cortés, on a wall in Boston, for 24 hours (see the webpage for details): Río Magdalena and Tribute to Fred Sandback.

https://www.bu.edu/humanities/pedestrian-space/

A Colloquium Lecture on interactions between Model Theory and the Foundations of Physics, by Boris Zilber,at UNAL, today (4 pm, Colombia time):

https://meet.google.com/cnc-ohtg-wiv