An international eye on Swiss news

ECONOMICS: Paul Piff, Daniel Stancato, Stéphane Côté, Rodolfo Mendoza-Denton, and Dacher Keltner, for amassing evidence that upper class people are more likely to snatch from a children's supply of candy, and engage in other kinds of unethical behaviour.

CHEMISTRY: Sanchari Banerjee, Nathan Coussens, François-Xavier Gallat, Nitish Sathyanarayanan, Jandhyam Srikanth, Koichiro Yagi, James Gray, Stephen Tobe, Barbara Stay, Leonard Chavas, and Subramanian Ramaswamy for discovering that milk proteins from vivaparous cockroaches packs three times as much energy as milk proteins from cows.

MEDICINE: The late Tokuji Unno, for the study called "How to Blow the Nose — An Aerodynamic Study of the Nose Blowing"

BIOLOGY: João Miguel Alves-Nunes, Adriano Fellone, Selma Maria Almeida-Santos, Carlos Roberto de Medeiros, Ivan Sazima, and Otavio Augusto Vuolo Marques, for gently stepping on various parts of various venomous snakes — repeatedly, at various temperatures at various times of day — to try to learn what does and what does not induce a venomous snake to bite a human.

PHYSICS: Randy Hurd, Zhao Pan, Tadd Truscott, Kaveeshan Thurairajah, for theoretical and experimental attempts to design a splash-free urinal.

TECHNOLOGY: Justin Puma, Zhen Yang, Evan Johnston, Zixin Zhang, Xiaoyi Lan, Lingzhi Zhang, Hongyu Hou, Zixin He, Ali Afify, Megan Creighton, Jianyu Li, and Changhong Cao, for pioneering the use of mosquito proboscises as high-resolution 3D printing nozzles, a process called "Necroprinting".

OLFACTION: Ilona Croy, Tomasz Frackowiak, Thomas Hummel, and Agnieszka Sorokowska, for collecting evidence that babies smell wonderful to their parents, but teenagers do not.

PEACE: Jaimie Arona Krems, Rebecka Hahnel-Peeters, Laureon Merrie, Keelah Williams, and Daniel Sznycer, for documenting that we value friends who are vicious to people we dislike.

SOIL SCIENCE: (Swiss research) Franz Bender, Marcel van der Heijden, Atlant Bieri, and Pia Viviani, for burying 1000 identical pairs of underpants in more than 25 countries, and then digging them up after two months.

Ig Nobel Prizes land in Switzerland on 3 September amid stricter US border policy (LINK) — 3 September 2026

— The prizes have been awarded for 35 years. Until now, they have always been handed out in the US. Researchers receive a genuine, albeit completely worthless, 10-trillion-dollar note from Zimbabwe. In 2008, the Ig Nobel Peace Prize was awarded to the Swiss National Ethics Commission and the entire Swiss population. The reason: Switzerland had enshrined in law the principle that plants, too, possess dignity.

Geneva Airport generates over CHF3 billion for regional economy: report (LINK) — 4 September 2026

— Of the CHF3.3 billion, nearly CHF2 billion went to the canton of Geneva, more than CHF550 million were generated in the neighbouring French region, and around CHF770 million in the rest of Switzerland.

Switzerland has worst summer potato harvest on record: nearly 20% less than the average over the last six years and 130,000 tonnes of potatoes had to be imported (LINK) — 3 September 2026

An AI capable of doubt can optimize scientific discovery: A team at EPFL have developed a framework that trains Large Language Models to create AI systems that can find the best possible setup or "recipe" for a scientific experiment (LINK) — 1 September 2026

— Bojana Rankovic and Philippe Schwaller at EPFL's Laboratory of Artificial Chemical Intelligence have developed a method that combines an LLM with a a Gaussian process, a kind of 'doubt detector", the probabilistic model commonly used in Bayesian optimization." Rather than asking the LLM to choose experiments directly ("direct prompting"), GOLLuM trains it using the Gaussian process's way of scoring uncertainty and performance of each option. Because of this, as the LLM learns from past experiments, it adjusts how it "organizes" the search space, the "master list" of every possible choice for an experiment.

— The researchers evaluated GOLLuM across 23 benchmark tasks covering organic synthesis, analytical and process chemistry, materials and catalysis, and molecular property optimization. Each optimization run began with ten low-performing observations, and the team used the same GOLLuM configuration across all benchmark tasks rather than tuning it separately for each problem, showing its versatility. Across the 23 benchmarks, GOLLuM ranked first on average, finding high-performing experimental conditions more consistently than Bayesian optimization with expert-designed descriptors.