education
Frank Calegari argues mathematics must change how it rewards research News
Frank Calegari’s guest essay on Terence Tao’s blog proposes changing hiring, journals and training as AI makes producing mathematical results cheaper.
K-means clustering: grouping examples around representative centers Lesson
K-means groups numerical examples around a chosen number of centers by minimizing squared distance, producing useful structure without proving the groups have meaningful labels.
Bootstrap resampling: estimating uncertainty by repeatedly rebuilding your sample Lesson
The bootstrap estimates sampling uncertainty by resampling observed units with replacement and recomputing a statistic, while preserving the structure that makes those units independent.
Francesco Maggi asks who will understand science when AI produces the proofs News
Francesco Maggi’s Tao-hosted essay argues that AI’s scientific value depends on shared access, human understanding and independent scrutiny.
SI.gov puts federal AI programs under a new name News
SI.gov is an official White House policy and education portal using the administration’s Super Intelligence terminology, rather than a demonstrated technical breakthrough.
Brief AI help lowered immediate independent performance in Berkeley experiments News
Berkeley-linked experiments found that direct AI assistance improved immediate task performance but lowered accuracy after the assistant was removed.
Google’s lawyer trial separates better AI-assisted work from better unaided judgment News
A randomized patent-lawyer trial found better assisted drafts, while the later unaided advantage was concentrated among experienced attorneys.
StudentBench finds AI and human GRE tutors produced equivalent immediate gains News
A randomized study of adult GRE learners found pooled AI tutoring and expert human tutoring produced statistically equivalent immediate score gains after one hour, while leaving child learning and long-term retention untested.
Equivalence testing: how to show two AI systems are close enough Lesson
Equivalence testing asks whether a measured difference is small enough to be practically unimportant, which is stronger and more useful than merely failing to find a statistically significant difference.
NYC Public Schools plans a grades 2K–8 moratorium on student-facing generative AI News
NYC Public Schools says it will implement a 2026–27 moratorium on student-facing generative AI in grades 2K–8, pairing it with limited approved high-school use and screen-time rules.
New York City banned generative AI for students through eighth grade News
New York City Public Schools announced a one-year moratorium on generative AI for students from pre-kindergarten through eighth grade on September 2, 2026, allowing only limited teacher-supervised pilots in high schools.
Students using generative AI got better homework grades and worse exam scores News
A 30-month study of 26,811 Chinese secondary students found that after adopting generative AI, homework scores rose about 18% and homework time fell about 30%, while closed-book monthly exam scores fell about 20%, with losses concentrated among students whose homework time dropped sharply.
Two Essays About AI and Your Brain, and One Actual Study News
A randomized experiment found that AI assistance impaired conceptual understanding, code reading, and debugging while delivering no significant average speed gain, which supports the concern behind this week's viral developer essays but not their proposed fix.
A Brown professor moved a final in person, and the near-perfect scores collapsed News
A Brown University economics professor who suspected AI cheating switched his take-home final to an in-person exam, and average scores fell by roughly half — a stark data point in the growing fight over what take-home 'mastery' now measures.
Smart glasses fed students live exam answers -- and schools have no idea how to stop it News
A cheating ring at Brown used AI-connected glasses to pipe real-time answers onto the lenses, and universities across Asia are seeing the same thing.
smol-kimi-k3 Tool
A runnable 49-million-parameter from-scratch trainer that deliberately keeps Kimi K3's unusual architectural ingredients rather than collapsing into a vanilla transformer, including the 3-to-1 KDA and gated MLA rhythm, ShortConv-based KDA, attention residuals, stable latent mixture of experts, and no positional encoding. Ships the recipe, a 10k-step checkpoint and verification checks, and is explicit that it is an architecture study rather than a capability claim.
GPTZero Tool
The widely used AI-writing detector (about 19M users) that estimates how likely a passage was machine-generated; being acquired by Superhuman to build a persistent authenticity layer.
DeepTutor Tool
Self-hostable tutoring workspace with knowledge bases, research, question generation, and learner memory; release 1.7.0 adds modular installation.