Mathematics, materials, history and scientific searches connected by a question rather than a common score.
These are editorial connections, not a chain of causation or a shared score of progress.
Half a second of real detector data. A rising trace. A new way to stay with the signal.
Watch / Explore · GW150914 →
Following and a reading bookmark stay on this device.
A five-year pledge could give U.S. researchers stronger tools to test ideas in medicine, energy, materials, and quantum computing. Access and experimental results will determine its impact.
Source published 2026-10-08 · EmergingNew company commitments and federal research awards promise practical help for research teams. The next test is what scientists can actually do with it.
Source published 2026-10-08 · EmergingInitially 722 manuscripts across 17 research areas; OpenAI’s current catalogue lists 721 manuscripts in 373 families after three October 7 withdrawals and two October 26 additions.
Source published 2026-10-06 · EmergingGoogle's AI & Economy ATLAS draws on 2,600 specialized models and a survey of more than 600 US and UK scientists, who reported almost seven hours per week saved while verification and experiments remained bottlenecks.
Source published 2026-09-15 · DemonstratedIn a University of Washington materials course, students examined patterns surfaced by Ai2 AutoDiscovery, checked literature, and decided whether each result was a useful question, coincidence, or data flaw.
Source published 2026-09-14 · EmergingNASA and IBM released a Lunar Foundation Model trained primarily on Lunar Reconnaissance Orbiter data for research tasks such as crater and volcanic-feature mapping and possible polar-ice analysis.
Source published 2026-09-10 · EmergingGoodfire used Ai2's open OLMo post-training stack to trace a known regression and inspect behavioral shifts.
Source published 2026-09-09 · DemonstratedUniversal Cell Embedding was trained across 36 million cells, more than 1,000 cell types, and eight species to create reusable representations for cell biology.
Source published 2026-07-08 · DemonstratedResearchers used an AI system called Robin to connect biological evidence and identify a candidate direction for dry age-related macular degeneration, then tested it in laboratory models.
Source published 2026-05-19 · ExperimentalGoogle DeepMind reported that AlphaEvolve improved a trained graph neural network's ability to find feasible solutions to the AC Optimal Power Flow problem from 14% to more than 88%.
Source published 2026-05-07 · DemonstratedResearchers trained Merlin on roughly six million images from 15,331 CT scans to support multiple abdominal imaging tasks.
Source published 2026-03-04 · DemonstratedAlphaGenome is a unified DNA-sequence model that takes up to 1 Mb of DNA and predicts thousands of functional genomic tracks, including gene expression, splicing, chromatin features, transcription-factor binding, and contact maps.
Source published 2026-01-28 · DemonstratedAnomalyMatch processed 99.6 million Hubble Legacy Archive cutouts and the study reports 1,339 anomalies, more than 800 of them previously undocumented, with experts reviewing the candidates.
Source published 2025-12-16 · DemonstratedGoogle DeepMind and Commonwealth Fusion Systems announced a partnership in which CFS uses the open-source TORAX plasma simulator in daily workflows to test and refine SPARC operating plans before the tokamak is turned on.
Source published 2025-10-16 · EmergingResearchers developed Deep Loop Shaping, a reinforcement-learning feedback controller tested on the LIGO Livingston Observatory to reduce control noise in a difficult mirror-control loop.
Source published 2025-09-04 · DemonstratedA multicenter randomized, controlled, double-blind superiority trial tested whether adding AI-detected spatio-temporal electrogram-dispersion targets to pulmonary-vein isolation improves treatment of drug-refractory persistent atrial fibrillation.
Source published 2025-02-14 · DemonstratedFunSearch combined language models with program evaluation to discover mathematical constructions.
Source published 2023-12-14 · DemonstratedGNoME used graph networks to search for energetically favorable crystal structures.
Source published 2023-11-29 · ExperimentalScientists used machine learning to help engineer an enzyme that breaks down PET plastic.
Source published 2022-04-27 · ExperimentalHistorians working with Ithaca improved the restoration of damaged ancient Greek inscriptions.
Source published 2022-03-09 · DemonstratedResearchers used reinforcement learning to control magnetic confinement in a research tokamak.
Source published 2022-02-16 · DemonstratedStarcloud says its Starcloud-1 satellite launched with an NVIDIA H100 and later trained nanoGPT in space.
Source published date unknown · EmergingShort source reports on announced work, early results and emerging uses.
This is a narrower, experimentally connected materials-design claim than a large database of predicted candidates. One synthesis does not establish usefulness in a device or manufacturing pathway.
A generative model for inorganic materials design · published 2025-01-16Captured 2026-09-06 · a lead, not a verified achievement.Open a record to inspect its original source, limitations and unresolved questions. Try the linked instruments where this question has one. Following keeps this question in your next reading view.
What moved? ↗