Google Research reported a wearable-based health-age model associated with subsequent hypertension, cardiovascular disease and type 2 diabetes in the All of Us cohort (22,748 participants). A companion world model simulated 12-week trajectories. In 490 blinded paired reviews, coaches preferred simulation-grounded plans in 71.4% of evaluations; physicians rated 97.6% of physiological claims accurate. This summary is limited to the November 11, 2026 institutional blog. The authors state all evaluations were offline and retrospective; clinical effectiveness and durable habit change require prospective trials.
Technology: Coachable health-age model, SensorWM action-conditioned world model, feasibility-constrained optimizer with a language model agent, multimodal CKM deviation model
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: retrospective or benchmark
- Tested: Retrospective clinical-risk modeling and blinded expert comparison of simulated coaching plans.
- Still unknown: Source checking does not establish independent confirmation. Review date: 2026-11-15.
- AI role: Built the health-age metric and multimodal CKM model from wearable and clinical markers, forecast physiological responses to habit changes with a world model, and filtered and converted simulated habit plans into coaching text with a language model agent.
Evidence sources: https://research.google/blog/using-wearables-to-enable-cardiovascular-kidney-metabolic-ckm-risk-profiling-for-personalized-coaching/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Prospective study of 196 people at five US sites, compared with 24-hour ambulatory blood pressure measurement: 65.8% sensitivity and 90.0% specificity for screening, and 78.4% and 90.9% in the confirmatory test (48 participants). A feasibility study; it does not replace the cuff, and many cases are missed.
Technology: AI-PPG-ACC-HTN (photoplethysmography and accelerometer)
Publication: Peer-reviewed paper · Test context: prospective or independent check. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: prospective or independent check
- Tested: A prospective study of 196 participants at five US sites compared smartwatch screening with 24-hour ambulatory blood pressure monitoring: 65.8% sensitivity and 90.0% specificity, with 78.4% and 90.9% in a confirmatory subgroup of 48.
- Still unknown: The authors call it early feasibility research; the smartwatch does not replace a cuff, many true cases are missed, the confirmatory subgroup is small, and generalizability is unclear.
- AI role: A 1-D convolutional network pretrained with self-supervised contrastive learning on PPG and accelerometer waveforms produced embeddings; a logistic regression output a hypertension probability.
Evidence sources: https://research.google/blog/detecting-signs-of-hypertension-using-wearable-sensors/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 4/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Nexus AI says it has deployed Google Research's chest X-ray AI as a certified diagnostic tool at about 40 facilities in six countries: more than 25,000 X-rays assessed by over 90 clinicians, with response times under 45 seconds. These are author-reported deployment figures, with no sensitivity or patient outcome data; WHO and FDA approvals are still in progress.
Technology: Foundation models for chest X-rays
Publication: Announcement only · Test context: Test context unspecified in annotation. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Announcement only
- Test type: not reported
- Tested: The post reports deployment figures (about 40 facilities, over 25,000 X-rays, over 90 clinicians) without sensitivity, specificity or patient outcome data from this deployment; it cites earlier peer-reviewed studies for performance.
- Still unknown: Deployment-level accuracy and outcomes are not given, WHO prequalification and FDA clearance are not granted, and the facility count across countries is unclear.
- AI role: The AI analyzed chest X-rays and flagged TB-suggestive findings for triage in under 45 seconds; clinicians made final decisions.
Evidence sources: https://research.google/blog/bringing-foundational-models-to-the-front-lines-of-tb-screening/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 1/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Prospective study of 385 pregnant women in Chicago and Nairobi: the model estimates gestational age from scans taken by minimally trained operators with a mean error of 4.18 days, non-inferior to the 4.48 days of expert sonographers. This is a non-inferiority result from only two sites, both relatively urban.
Technology: Ultrasound model for estimating gestational age
Publication: Peer-reviewed paper · Test context: prospective or independent check. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: prospective or independent check
- Tested: A prospective multi-center study of 385 pregnant women at two sites (Chicago and Nairobi) compared AI gestational-age estimates from blind sweep scans by minimally trained operators with expert sonographer biometry: mean error 4.18 vs 4.48 days.
- Still unknown: The source does not state the non-inferiority margin or confidence intervals; only two relatively urban sites and one commercial probe were used, and clinical trust and confidence safeguards are not validated.
- AI role: The model estimated gestational age from blind sweep ultrasound scans and ran live on mobile devices in clinics.
Evidence sources: https://research.google/blog/empowering-frontline-maternal-care-with-ai-assisted-ultrasound/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 4/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
AI agents designed DNA aptamers and protein binders, which were then tested in the lab over multiple rounds. Target success rates reached 33.5% for myoglobin and 42.1% for α-thrombin (98.5% specificity). The post summarizes three arXiv papers whose peer review is unconfirmed; it also reports initial failures.
Technology: AI agents with laboratory robots
Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: direct test
- Tested: Agents ran three design-test-iterate rounds of DNA aptamers against myoglobin and alpha-thrombin in the lab, reaching 33.5% and 42.1% hit rates with 98.5% specificity for thrombin; protein binders against FAP were reported at single-digit nanomolar affinity.
- Still unknown: Peer review of the cited arXiv papers is unconfirmed; there are two protein targets and three rounds; initial rounds failed, and protein expression adds steps to each loop.
- AI role: Agents designed and revised sequences across rounds and adjusted design rules after reading wet-lab data, under human oversight of protocols and robot runs.
Evidence sources: https://research.google/blog/grounding-ai-agents-in-biological-reality-building-blocks-for-autonomous-wet-labs/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
The PDFM model increased the explained variance in MMR vaccination coverage from 16% to 22% across 146 counties on the US–Canada border, and improved weekly correct cholera-risk-area rankings in the DRC from 1.78 to 2.10. Gains for cardiovascular mortality and postpartum depression were small or not significant.
Technology: Population Dynamics Foundation Model (PDFM), TimesFM
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: retrospective or benchmark
- Tested: Five partner-led evaluations compared PDFM embeddings with baselines: MMR explained variance rose from 16% to 22% across 146 border counties and cholera Precision@5 improved from 1.78 to 2.10; cardiovascular gains were not significant.
- Still unknown: Several results were non-significant or small, cholera gains were not significant at one or two weeks, and the authors cite static embeddings as a limitation.
- AI role: PDFM combines search trends, mobility, built-environment and environmental data into location embeddings used without task fine-tuning; for dengue it was paired with TimesFM 2.0.
Evidence sources: https://research.google/blog/earth-ais-planetary-geospatial-foundation-models-for-global-public-health/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
MAPL-EMIT used a vision transformer trained on 3.6 million synthetic plumes embedded in EMIT radiance data to map methane sources. Beyond synthetic tests, the paper abstract reports recovery of 84% of known annotated plume complexes across 1,084 EMIT granules. Comparisons with airborne data, landfills and controlled releases supported previously uncaptured sources. This is source-reported validation, not an independent emissions audit.
Technology: MAPL-EMIT (end-to-end vision transformer on EMIT hyperspectral radiance)
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: Synthetic tests plus known real plumes, coincident airborne observations, landfill comparisons and controlled-release data.
- Still unknown: Review scope: complete indexed abstract and metadata; full uncertainty budgets and individual source emissions were not assessed.
- AI role: An end-to-end vision transformer jointly estimated methane enhancements, plume outlines and source locations from hyperspectral radiance.
Evidence sources: https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI%3A%2210.1073%2Fpnas.2612145123%22&resultType=core&format=json
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Google's Empirical Research Assistance (ERA), a Gemini-based tree-search system, reaches expert-level performance on six benchmarks spanning genomics, public health, satellite imagery, neuroscience, time-series forecasting and mathematics. Its state-level U.S. hospital admission forecasts for flu, COVID-19 and RSV rank at or near the top of CDC leaderboards. Results are computational and checked against public benchmarks and leaderboards, with no wet-lab validation; exact benchmark scores were not visible in the sources read.
Technology: Empirical Research Assistance (ERA), Gemini-based tree search
Family Gemini: profile · models page
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: ERA-generated methods and models were compared against human-developed methods on a public leaderboard for bioinformatics and against the CDC ensemble and other individual models for COVID-19 hospitalization forecasts.
- Still unknown: Not reported.
- AI role: A Gemini-based large language model with tree search generated scientific software, including 40 single-cell data analysis methods and 14 forecasting models.
Evidence sources: https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI:10.1038/s41586-026-10658-6&resultType=core&format=json
The legacy label “Not reported” does not distinguish information absent from reviewed material from information not assessed. It is not a scientific finding of absence.
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
For insertions and deletions, it reaches 90% with Illumina data versus 80% for the best other tool, and over 80% with PacBio versus under 50%. In eight pediatric leukemia samples, it recovered known variants and flagged ten new ones that have not been independently validated. The tool and CASTLE dataset are open.
Technology: DeepSomatic (convolutional neural network)
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: DeepSomatic was benchmarked against SomaticSniper, MuTect2, Strelka2 and ClairS on the CASTLE cell-line dataset with held-out chromosomes and genomes, across Illumina, PacBio and other platforms.
- Still unknown: The benchmark uses research cell lines rather than patient tumors, the glioblastoma result rests on one sample, and the 10 new leukemia variants are not stated to be validated.
- AI role: A convolutional neural network applied to sequencing data converted to images separated germline, somatic and error signals to call tumor variants.
Evidence sources: https://research.google/blog/using-ai-to-identify-genetic-variants-in-tumors-with-deepsomatic/ · https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI:10.1038/s41587-025-02839-x&resultType=core&format=json
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 59 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
In a blinded remote OSCE with 149 case scenarios, AMIE and primary care physicians held text consultations with simulated patients. Specialist physicians rated AMIE superior on 30 of 32 axes and patient actors on 25 of 26, and non-inferior on the rest. The evidence comes from simulated scenarios, not real patients, and the authors state that further research is needed before real-world use.
Technology: AMIE (Articulate Medical Intelligence Explorer), a large language model agent for diagnostic dialogue
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: direct test
- Tested: A randomized, double-blind crossover study of text consultations with patient-actors across 159 case scenarios compared AMIE with 20 primary care physicians, rated by specialist physicians and patient actors on several axes.
- Still unknown: The authors note the unfamiliar synchronous text-chat format and state that further research is needed before real-world use.
- AI role: AMIE, an LLM-based diagnostic dialogue system trained by self-play with automated feedback, conducted the consultations.
Evidence sources: https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI:10.1038/s41586-025-08866-7&resultType=core&format=json
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 824 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 4/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A system based on Gemini 2.0 that generates hypotheses and research proposals. In the lab, three drugs it suggested for repurposing against acute myeloid leukemia reduced cancer-cell viability, and all suggested treatments for liver fibrosis showed activity. Expert assessments covered only 11 targets; fibrosis results awaited a Stanford report.
Technology: AI co-scientist (Gemini 2.0, multi-agent)
Family AI Co-Scientist: profile · models page
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: direct test
- Tested: Elo auto-evaluation over 203 research goals; 15 expert-curated goals; blinded experts rated 11 goals (preference rank 2.36). In vitro on four AML cell lines: binimetinib, pacritinib, cerivastatin reduced viability; KIRA6 IC50 10 nM in KG-1a versus 180 nM in TK6. Seven combinations tested.
- Still unknown: The paper calls validation preliminary and in vitro only. KIRA6 was far less effective in MOLM-13 and HL-60. Expert evaluation was small and subjective; Elo is not ground truth. Liver fibrosis and AMR details are summarized only.
- AI role: A Gemini-based multi-agent system generated, debated and ranked hypotheses; experts selected candidates and prioritized lab experiments.
Evidence sources: https://doi.org/10.1038/s41586-026-10644-y
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
The new model, trained on 15,980 stations (up from 5,680), gives seven-day reliability comparable to the previous model's five-day forecast. Flood Hub covers over 100 countries and about 700 million people (up from 460 million). About 250,000 virtual points are unverified and intended for experts only; flash, urban, and coastal floods remain out of scope. The linked March 2024 Nature paper describes the predecessor model, also covered by the catalog's March entry; the November performance and coverage update is reported in this Google Research announcement.
Technology: Hydrological forecasting model
Publication: Announcement only · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Announcement only
- Test type: retrospective or benchmark
- Tested: Gauged basins were checked with standard cross-validation, and ungauged locations were checked against Sentinel-1 radar inundation images; reliability was compared using Nash-Sutcliffe efficiency distributions across 3,079 basins.
- Still unknown: Ungauged validation covers only a fraction of locations, flash, urban and coastal floods are out of scope, and virtual-gauge forecasts are unverified.
- AI role: An LSTM network takes meteorological time series and static catchment attributes, adds input from DeepMind's medium-range forecast model, and outputs a probabilistic streamflow distribution.
Evidence sources: https://research.google/blog/a-flood-forecasting-ai-model-trained-and-evaluated-globally/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 580 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
At 1.4° resolution, the model outperformed ECMWF-ENS in 95% of cases for forecasts from 2 to 15 days; at 2.8°, its mean 1980–2020 temperature error was 0.25 °C versus 0.75 °C for atmosphere-only models. It is more than 3,500 times faster than X-SHiELD on a TPU. It models only the atmosphere and is not yet a complete climate model.
Technology: NeuralGCM (differentiable dynamical core and neural networks)
Family NeuralGCM: profile · models page
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: Evaluated with the WeatherBench 2 benchmark against ECMWF-ENS (better in 95% of 2-15 day cases at 1.4 degrees) and against AMIP climate baselines for 1980-2020 temperature (0.25 vs 0.75 degrees C error).
- Still unknown: The model covers only the atmosphere and is not yet a complete climate model; oceans and the carbon cycle are future goals.
- AI role: A neural network learns small-scale physics such as cloud formation, replacing hand-built parameterizations, coupled to a physics-based fluid dynamics solver.
Evidence sources: https://research.google/blog/fast-accurate-climate-modeling-with-neuralgcm/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 780 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
The model produces 256 members at 2° resolution every three minutes on a TPU and matches or exceeds the operational GEFS ensemble (31 members) on rank histograms, RMSE and CRPS. For the Lisbon heatwave (July 2022), with 16,384 members, none of GEFS’s 31 members was as warm as observed. One extreme case; depends on physics-based forecasts as seeds; published in Science Advances.
Technology: SEEDS (diffusion model)
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: The ensemble was compared with the operational GEFS (31 members) using rank histograms, RMSE and CRPS, and with a 16,384-member ensemble for the Lisbon heatwave of July 2022.
- Still unknown: The heatwave comparisons are single examples, the model depends on physics-based seed forecasts, and the page has no explicit limitations section.
- AI role: A denoising diffusion model generates large weather-forecast ensembles from one or two operational forecasts.
Evidence sources: https://research.google/blog/generative-ai-to-quantify-uncertainty-in-weather-forecasting/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 172 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A model trained on 5,680 gauges improved average global forecast reliability from zero to five days: forecasts at days 4–5 were similar to or better than day-zero forecasts from GloFAS v4. Flood Hub covers more than 80 countries. F1 comparison across 2,092 basins for two-year return-period events; flash and urban floods remain to be covered.
Technology: Hydrological forecasting model
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: F1 scores were compared with GloFAS v4 for 2-year return-period events across 2,092 watersheds (2014-2023); reliability at 4-5 day lead times was similar to or better than GloFAS at day zero.
- Still unknown: Coverage must expand to more locations, flash and urban floods are not yet covered, and the gain is average-level rather than uniform.
- AI role: Two sequential LSTMs with a mixture density output produce probabilistic streamflow forecasts, trained on 5,680 gauges.
Evidence sources: https://research.google/blog/using-ai-to-expand-global-access-to-reliable-flood-forecasts/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 580 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
The model forecasts precipitation, temperature, wind and dew point up to 24 hours ahead at 1–4 km resolution, learning from about 1,000 stations, and outperforms HRRR and ENS in several regions. It operates in Google Search and the app in the United States and 27 European countries, with precipitation forecasts up to 12 hours. Preprint, no numerical values in the text and one case study.
Technology: MetNet-3
Family MetNet: profile · models page
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: retrospective or benchmark
- Tested: MetNet-3 was benchmarked against HRRR and ENS over the contiguous US for forecasts up to 24 hours, scored with CRPS; a single January 2019 case study is also shown.
- Still unknown: The post has no limitations section; it notes operational coverage limited to the contiguous US and parts of Europe, hyperlocal wind results depend on test-station input, and the case study is a single example.
- AI role: A neural model forecasts precipitation, temperature, wind and dew point up to 24 hours ahead, learning from about 1,000 stations.
Evidence sources: https://arxiv.org/abs/2306.06079 · https://research.google/blog/metnet-3-a-state-of-the-art-neural-weather-model-available-in-google-products/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
In a trial with American Airlines and Breakthrough Energy, pilots flew 70 flights over six months using AI forecasts to avoid forming contrails, reducing them by 54% compared with flights without the forecasts. Flights that avoided contrails used 2% more fuel; across an airline’s flights, the impact could fall to 0.3% if only a small share need rerouting.
Technology: AI contrail-formation forecasts and satellite imagery
Publication: Announcement only · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Announcement only
- Test type: direct test
- Tested: Google, American Airlines and Breakthrough Energy flew 70 pilot flights over six months using AI forecasts to avoid contrail-forming altitudes; satellite imagery reported a 54% contrail reduction.
- Still unknown: The article does not state a control group, the baseline for the 54% figure, statistical methods or peer review; fuel impact (2% more fuel, 0.3% airline-wide) is an estimate.
- AI role: AI contrail forecasts guided pilots to avoid altitudes likely to form contrails.
Evidence sources: https://blog.google/technology/ai/ai-airlines-contrails-climate-change/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A consortium of 119 scientists from 60 institutions published in Nature a pangenome combining assembled genomes from 47 people of diverse ancestries, rather than one person. Google adapted DeepVariant to identify variants and DeepConsensus to correct sequencing errors; a long-read assembler achieved over 99.999% final accuracy. This is a draft: benefits for diagnosis and treatment remain prospective.
Technology: DeepVariant and DeepConsensus
Family DeepConsensus: profile · models page
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: retrospective or benchmark
- Tested: The draft pangenome from 47 assemblies was evaluated for accuracy at structural and base-pair levels, and short-read analysis was compared with GRCh38-based workflows (34% fewer small-variant errors).
- Still unknown: The abstract states no explicit limitations; the blog says this is a draft and benefits for diagnosis and treatment remain prospective.
- AI role: Google adapted DeepVariant to call variants and DeepConsensus to correct sequencing errors in the pangenome assembly.
Evidence sources: https://blog.google/innovation-and-ai/technology/health/first-pangenome-reference-nature-paper-ai/ · https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI:10.1038/s41586-023-05896-x&resultType=core&format=json
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 1258 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A model using geostationary imagery tracks fire perimeters with updates every 10–15 minutes and, according to the page, is more accurate than similar products and the previous model based on NOAA GOES-16/17 data; shown in Search and Maps for large fires in the United States, Mexico, much of Canada and Australia. Resolution is 2 km; clouds can hide fires; figures appear only in an image.
Technology: Model on geostationary satellite imagery
Publication: Announcement only · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Announcement only
- Test type: retrospective or benchmark
- Tested: Fire perimeters were compared with official fire scars using precision, recall and distance metrics; a comparison with an earlier GOES-16/17 model and a 2022 US test set unavailable during training are also reported.
- Still unknown: The post states limitations: smoke and clouds hide fires, resolution is about 2 km and coarser away from nadir, and coverage is limited to large fires in the listed regions; no paper is cited.
- AI role: A model using geostationary satellite imagery tracks fire perimeters with updates every 10-15 minutes.
Evidence sources: https://research.google/blog/real-time-tracking-of-wildfire-boundaries-using-satellite-imagery/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A high-throughput optical screen and data-science models identified a Co–Ta–Sn composition region whose optical trends indicated phase behavior beyond simple mixtures. Follow-up X-ray diffraction verified a rutile substitutional-alloy oxide, (Sn,Co,Ta)O₂. Composition tuning produced optical transparency and oxygen-evolution electrocatalytic activity with stability in strong-acid electrolyte, as reported by the authors.
Technology: Physics-aware data-science phase-diagram and emergent-property models
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: direct test
- Tested: A high-throughput optical screen and data-science models identified a Co–Ta–Sn composition region whose optical trends indicated phase behavior beyond simple mixtures.
- Still unknown: The phase inference initially came from optical data and the authors discuss potential false positives from extrinsic film morphology; the selected Co–Ta–Sn rutile phase was followed up and verified by XRD. The optical screen identifies promising regions, not all possible phases, and may miss phases without emergent optical signatures. Reported oxygen-evolution performance and acid stability are laboratory electrocatalysis results; no commercial device or scale-up is established.
- AI role: Models analyzed high-throughput optical absorption measurements across 376,752 compositions in 108 three-cation oxide systems. They inferred candidate phase diagrams and flagged optical trends inconsistent with lower-order phase mixtures, guiding the Co–Ta–Sn follow-up composition and XRD experiments that verified a rutile alloy phase; this screening also guided catalytic characterization.
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC8449358/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Citations: at percentile 88 of works of the same type, year and subfield (OpenAlex work metrics, 38 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
A sample of about 1 mm³ of human cortex was imaged at 4 nm across about 5,300 sections and automatically reconstructed: 1.4 petabytes, tens of thousands of neurons, 130 million annotated synapses, and 104 manually corrected cells. The sample is about one-millionth of a brain; reconstruction is imperfect and initial findings are preliminary.
Technology: Automated segmentation of electron microscopy images
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: retrospective or benchmark
- Tested: The automated reconstruction was checked by manual proofreading of about 104 cells; the abstract says connections were validated.
- Still unknown: Automated reconstruction is imperfect, the sample is about one-millionth of a brain, and findings are preliminary.
- AI role: Automated machine-learning pipeline reconstructed neurons and synapses from electron microscopy images.
Evidence sources: https://research.google/blog/a-browsable-petascale-reconstruction-of-the-human-cortex/ · https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI%3A10.1101/2021.05.29.446289&resultType=core&format=json
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Neural models trained on particle-display affinity-screening data guided sequence mutation and candidate nomination, followed by synthesis and experimental affinity testing. The authors report a 23-nucleotide truncation of candidate G13 with measured dissociation constant 1.5 nM, compared with about 8 nM for the full-length parent under their assay. This is a binding result, not demonstrated diagnostic or therapeutic efficacy. The screening set is not 187,499 independently validated agents; sequence-level prior art remains open.
Technology: Neural-network models trained on particle-display affinity data
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: direct test
- Tested: The study reports particle-display screening and measured NGAL binding, including full Kd for truncated and parent G13.
- Still unknown: Binding assay only; no diagnostic/therapeutic outcome; one holdout set used for tuning and sequence novelty requires further review.
- AI role: Trained neural models extrapolated from experimentally measured sequences to nominate and iteratively refine higher-affinity DNA aptamers and truncations.
Evidence sources: https://www.nature.com/articles/s41467-021-22555-9
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 16 Nov 2026. Review scope: Primary publisher indexed full text, article dates, methods/results, affiliations and disclosed limitations..
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
Models trained on DNA-encoded-library selection data prioritized roughly 2,000 compounds for biochemical testing across three targets. The study reports active and potent hits outside the training libraries. Human filtering and chemical review contributed to selection. These are biochemical hits, not established safe or effective drugs; the authors disclose commercial interests and related patent applications.
Technology: Graph-convolutional neural networks and random forests
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Peer-reviewed paper
- Test type: direct test
- Tested: Newly selected compounds were tested in biochemical assays.
- Still unknown: Cellular efficacy, selectivity, safety and clinical benefit remain unestablished.
- AI role: Graph-convolutional neural networks and random forests supported the stated study; experimental and interpretive work remained with researchers.
Evidence sources: https://accio.github.io/AMIDD/assets/2020/13-14/McCloskey-2020-ML-DELT.pdf
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 16 Nov 2026. Review scope: Author-hosted published-article PDF abstract, selected experimental Results, affiliations and disclosures; prior PubMed online date retained..
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.
The model forecasts precipitation at 1 km resolution and 2-minute intervals up to 8 hours ahead from radar and satellite, without physical equations. According to the page, it outperforms NOAA’s operational HRRR model up to about 7–8 hours, in seconds rather than about an hour. US coverage only, author-only comparisons, a preprint, and no stated operational use.
Technology: MetNet (neural network)
Family MetNet: profile · models page
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
Result, AI contribution and limits
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
- Publication status: Preprint
- Test type: retrospective or benchmark
- Tested: Evaluated on a precipitation-rate forecasting benchmark against NWP (HRRR) and an optical-flow baseline, scored by F1 at 1.0 mm/h; outperforms NWP at forecasts up to about 7 to 8 hours.
- Still unknown: Demonstrated only for the continental US; the authors call it a small stepping stone; comparisons are author-run.
- AI role: Deep neural network with axial self-attention forecasting precipitation probabilities from radar and satellite inputs, without physical equations.
Evidence sources: https://research.google/blog/a-neural-weather-model-for-eight-hour-precipitation-forecasting/ · https://arxiv.org/abs/2003.12140
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: blog. Paper: linked separately.
Review scope and date are not recorded; source checked does not establish full-text review.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.
Follow-up and relationships
No follow-up recorded; this is not a claim that no follow-up exists.
Follow-up review scope and date are not recorded.
No source-supported relationships recorded.