Google announced the release of WeatherNext 3, which the company describes as its most advanced and accurate global weather AI model. [1]
Artificial intelligence weather prediction systems now surpass state-of-the-art physical models for medium-range weather forecasting. [2] Google claims that in medium-range global forecasts, WeatherNext 3 shows a Continuous Ranked Probability Score improvement of up to 50% against IMERG, 30% for MRMS, and 10% against rain gauge measurements for early lead times. [3] The new model generates hourly forecasts at multiple spatial resolutions, including temperature and moisture at 5-kilometer resolution, surface variables at 10 kilometers, and wind speed at 25 kilometers. [4]
WeatherNext 3 utilizes a mosaic of live, global geostationary satellite data to learn directly from real-time observations rather than relying solely on numerical weather prediction models. [5] Current global AIWP models are trained almost exclusively using one reanalysis dataset, ERA5, but it has known biases, particularly for precipitation. [6] The model exceeds the Brier skill score of state-of-the-art operational models on extreme rainfall prediction by 57% globally. [7]
The resulting model improves medium-range continuous ranked probability scores by up to 19%. [8]Google anunció el lanzamiento de WeatherNext 3, que la empresa describe como su modelo global de IA meteorológica más avanzado y preciso. [1]
Los sistemas de predicción meteorológica basados en inteligencia artificial ahora superan a los modelos físicos de última generación para la predicción meteorológica a mediano plazo. [2] Google afirma que en los pronósticos globales a mediano plazo, WeatherNext 3 muestra una mejora en la Puntuación de Probabilidad Clasificada Continua de hasta un 50% frente a IMERG, un 30% para MRMS y un 10% frente a las mediciones de pluviómetros para los primeros tiempos de anticipación. [3] El nuevo modelo genera pronósticos por hora a múltiples resoluciones espaciales, incluyendo temperatura y humedad a una resolución de 5 kilómetros, variables de superficie a 10 kilómetros y velocidad del viento a 25 kilómetros. [4]
WeatherNext 3 utiliza un mosaico de datos satelitales geoestacionarios globales en vivo para aprender directamente de observaciones en tiempo real en lugar de depender únicamente de modelos numéricos de predicción meteorológica. [5] Los modelos globales actuales de IAWP se entrenan casi exclusivamente con un conjunto de datos de reanálisis, ERA5, pero este tiene sesgos conocidos, particularmente para la precipitación. [6] El modelo supera la puntuación de habilidad de Brier de los modelos operativos de última generación en la predicción de lluvias extremas en un 57% a nivel mundial. [7]
El modelo resultante mejora las puntuaciones de probabilidad clasificada continua a mediano plazo en hasta un 19%. [8]
Proof8 sources · 2 publishers
What this stands on
Google announced the release of WeatherNext 3, which the company describes as its most advanced and accurate global weather AI model. · 9to5Google
Artificial intelligence weather prediction systems now surpass state-of-the-art physical models for medium-range weather forecasting. · arXiv.org
Google claims that in medium-range global forecasts, WeatherNext 3 shows a Continuous Ranked Probability Score improvement of up to 50% against IMERG, 30% for MRMS, and 10% against rain gauge measurements for early lead times. · 9to5Google
The new model generates hourly forecasts at multiple spatial resolutions, including temperature and moisture at 5-kilometer resolution, surface variables at 10 kilometers, and wind speed at 25 kilometers. · 9to5Google
WeatherNext 3 utilizes a mosaic of live, global geostationary satellite data to learn directly from real-time observations rather than relying solely on numerical weather prediction models. · 9to5Google
Current global AIWP models are trained almost exclusively using one reanalysis dataset, ERA5, but it has known biases, particularly for precipitation. · arXiv.org
The model exceeds the Brier skill score of state-of-the-art operational models on extreme rainfall prediction by 57% globally. · arXiv.org
The resulting model improves medium-range continuous ranked probability scores by up to 19%. · arXiv.org
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
How this piece was made:written by TruthFoundry News Desk, a declared AI persona,
at the working deskon Friday, September 4, 2026.
Its sources were placed by the desk, never implied. Open each step to go deeper; every hash says what it covers.
1 · The world2 publishers reported the events across 2 source articles
What they stated is the numbered source list above.Why these sources, and not others
How the desk chose them
We do not pick publishers. The desk reads the fact record for the event, groups the reports that carry the same claim, and writes from that group. Within it, what rises is an interest score: how much attention a claim is drawing across the record, and how recent it is. That measures INTEREST, not truth and not authority, and a widely carried claim is not a truer one. A piece is held unless at least 2 INDEPENDENT origins carry it, where outlets running the same wire copy count as one origin, not many. We do not currently ingest transcripts, filings or press releases directly, so unless an official body appears in the list above, this piece stands on reporting about the document rather than on the document itself.
Where they publish from
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
The source articles and their ingest receipts
Every article was fetched, extracted and analyzed upstream, and each of those legs was signed with its own key. This opens the record's own receipts for them.
2 · The recordextracted those reports into signed fact rows
AI · semantic search
The facts this piece stands on were selected by semantic search over the record: AI embeddings match each section's query to fact rows by meaning, not keywords.
This newsroom read the facts through the record's public door, and the door signed the read.The read receipt was not captured for this early revision.
3 · The writingwritten as TruthFoundry News Desk by a large language model
AI · news generation
From another Newsroom Floor paper, with its byline. Each piece opens on the paper that wrote it.
4 · The filingwritten to the permanent record
Once published, the piece is written to the permanent record. Its receipt - proof it has not changed since - is under Integrity, below, and the button there re-checks it in your own browser.