NSF AI Institute on Trustworthy AI for Weather, Climate, and Coastal Oceanography (AI2ES)

What NSF AI Institute on Trustworthy AI for Weather, Climate, and Coastal Oceanography (AI2ES) does

AI2ES is a non-profit, multi-institutional research institute funded by the U.S. National Science Foundation under Grant No. RISE-2019758, established in 2020 and headquartered at the University of Oklahoma under Director Amy McGovern. The institute develops and deploys artificial intelligence and machine learning models for weather forecasting, climate modeling, and coastal oceanography, with explicit emphasis on trustworthiness, explainability, and uncertainty quantification. Core technical assets include the FogNet deep learning model for coastal fog prediction (9-24 hour lead time), OceanNet with Fourier Neural Operators for ocean circulation (up to 120-day forecasts), MILES-GUESS for evidential deep learning, Ptype-Physical for winter precipitation classification, Cold-Stunning Prediction for sea turtle conservation (120-hour lead time), and the Semaphore platform that operationalizes 55 production AI models as of March 2024. The institute operates as a consortium of 8+ universities plus NSF NCAR, with 12 core staff supported by 30+ graduate students and 40+ undergraduates.

AI2ES does not sell commercial products. All code is released under Creative Commons 0 open-source licensing and distributed via GitHub, with revenue arriving exclusively through competitive NSF grants. The recent $3M+ in NSF Collaborations in Artificial Intelligence and Geosciences (CAIG) funding for four new projects demonstrates continued federal support. The go-to-market is event-driven and community-led: dedicated sessions at American Meteorological Society conferences, bi-weekly site-wide seminars, peer-reviewed publications, annual #CODE_IT STEM camps, and short courses and tutorials. Primary beneficiaries are government weather and transportation agencies (NWS, NOAA, NYSDOT, Nueces County Coastal Parks, Texas Parks and Wildlife), academic researchers, and students entering the AI for Earth-system science workforce.

AI2ES differentiates itself through an integrated multi-disciplinary approach combining AI, atmospheric science, oceanography, and risk communication under a single coordinated research agenda, with 15 industrial and government partners co-developing solutions. The institute contributed to the World Meteorological Organization's United in Science 2024 Report, positioning its leadership in the global trustworthy AI policy discourse. Production deployments include 2,400 NYSDOT traffic cameras for road weather monitoring and Texas coastal warning systems for marine cold-stunning events.

Name

NSF AI Institute on Trustworthy AI for Weather, Climate, and Coastal Oceanography (AI2ES)

Legal name

NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography

Website

https://ai2es.org

Company type

Private

Operating status

Operating

Headcount range

11–50 employees

Short description

AI2ES is an NSF-funded, multi-institutional research institute that develops open-source AI/ML models for weather, climate, and coastal oceanography, serving government weather and transportation agencies, academic researchers, and students through open-source tools and research-to-operations partnerships.

Ownership category

akta.pro rank

Product category

AI for Earth System Science

NAICS

Software Publishers (5132), Computer Systems Design and Related Services (54151), Software Publishers (513210)

SIC

Services-Prepackaged Software (7372), Services-Computer Programming Services (7371)

akta.pro industry

PRIMARYHDAAAKAL

Audit, Explainability & Accountability Tooling (traceability, reporting)

Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/AI Security, Privacy & Model Governance

SECONDARYHDAEANAF

AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety)

Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise

SECONDARYHDAEANAA

End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management)

Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise

SECONDARYBPAEAHAF

AI/ML Engineering & Model Development Services

Business & Public Services/IT & Digital Services (Managed Services, Systems Integration)/Data, Analytics & AI Services

SECONDARYHDAAAMAC

Bias, Fairness & Non-Discrimination Testing

Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/AI Ethics & Compliance Solutions

Keywords

• AI weather forecasting
• Climate modeling research
• Coastal oceanography AI
• Trustworthy machine learning
• Earth system science

Headquarters

HQ city

Norman (University of Oklahoma)

HQ country

United States

HQ region

North America

Offices9 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Operations, Infrastructure, Marketing or Sales

Revenue model

1. NSF Grant Funding: Primary funding from the U.S. National Science Foundation under Grant No. RISE-2019758. Additional funding from Collaborations in Artificial Intelligence and Geosciences (CAIG) program totaling over $3 million for 4 new projects.

Go-to-market motion2 records

Distribution channels3 records

Marketing channels6 records

Core offering

AI2ES is a multi-institutional NSF-funded research institute that develops open-source AI/ML models and Python software tools for weather forecasting, climate modeling, and coastal oceanography. Its offerings include FogNet (coastal fog prediction), OceanNet (ocean circulation forecasting), Semaphore (operational AI deployment platform), MILES-GUESS (uncertainty quantification), Ptype-Physical (winter precipitation type classification), and the CREDIT Earth Digital Intelligence Twin. All software is released free of charge under Creative Commons 0 licensing to support research-to-operations transitions with NCAR, NOAA, NWS, NYSDOT, and The Weather Company.

Product overview

AI2ES is a research institute (not a commercial product company) that develops open-source AI/ML tools and models for weather, climate, and coastal oceanography applications. The core offerings include: Semaphore (a Python platform for semi-automating AI model operationalization), FogNet (deep learning for coastal fog prediction), OceanNet (Fourier Neural Operator-based ocean circulation models), MILES-GUESS (evidential deep learning for uncertainty quantification), Ptype-Physical (winter precipitation type classification), and CREDIT (Earth Digital Intelligence Twin). These tools are primarily open-source and designed for research-to-operations deployment in collaboration with agencies like NCAR, NOAA, and The Weather Company.

Differentiator

Problem solved

Functional benefit

Brands

• FogNet: A deep learning model that combines satellite sea surface temperatures and numerical model outputs to predict the onset of fog with 9 to 24 hours advance warning.
• OceanNet
• Semaphore
• Ptype-Physical

Products and services

• Semaphore A Python application that semi-automates the operational deployment of AI models, downloading data, running .H5 models, post-processing outputs, and saving results; designed for research-to-operations transition.
• FogNet A deep learning model combining satellite sea surface temperatures and numerical model outputs to predict coastal fog onset with 9–24 hour advance warning for coastal operations, aviation, and transportation.
• OceanNet A deep convolutional neural network with Fourier Neural Operators for regional ocean circulation prediction, providing superior forecasts compared to dynamical models for up to 120 days.
• MILES-GUESS Python package providing Keras and PyTorch implementations of evidential deep learning models for analysis of aleatoric and epistemic uncertainties in Earth system science.
• Ptype-Physical Python code for generating machine learning estimates of winter precipitation type from vertical atmospheric profiles, with uncertainty quantification for winter weather prediction.
• CREDIT (Community Research Earth Digital Intelligence Twin) A digital intelligence twin framework for Earth system modeling co-developed with NCAR, combining ML models for numerical weather prediction with uncertainty quantification.

Quantifiable outcome

• 130+ peer-reviewed articles published
• +5 more outcomes

Named customers5 records

Segments4 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability9 records

Feature6 records

Partnerships

13 partnerships are on record, tiered core, flagship and minor.

• National Center for Atmospheric Research (NCAR)coreStrategic or Co-development PartnerDeep collaboration with NCAR including shared researchers (David John Gagne, Julie Demuth, Mariana Cains, John Schreck), MILES project development, SIParCS internship program hosting, and co-development of CREDIT (Community Research Earth Digital Intelligence Twin).
• The Weather Company (IBM)coreStrategic or Co-development PartnerCollaboration on operational AI prediction models for cold-stunning events and coastal applications. The Weather Company provides industry expertise for transitioning research models to operations.
• National Oceanic and Atmospheric Administration (NOAA)coreStrategic or Co-development PartnerNOAA collaboration includes workforce development initiatives, joint workshops on AI in environmental sciences, and research-to-operations support through Hazardous Weather Testbed.
• World Meteorological Organization (WMO)flagshipStrategic or Co-development PartnerAI2ES Director Amy McGovern participated in drafting the WMO United in Science 2024 Report, contributing expertise on AI/ML for weather forecasting and ethical AI development.
• University of OklahomacoreStrategic or Co-development PartnerPrimary institutional host for AI2ES Director Amy McGovern and core research team. Partners with National Weather Center REU program for undergraduate research.
• Texas A&M University – Corpus ChristicoreStrategic or Co-development PartnerLead institution for coastal research including Philippe Tissot and the Conrad Blucher Institute. Hosts operational water temperature and water level prediction systems.
• Del Mar CollegecoreStrategic or Co-development PartnerCommunity college partner for AI curriculum development including first-of-its-kind AI Occupational Award. Hosts annual #CODE_IT STEM camp for 30+ middle school students.
• Colorado State UniversitycoreStrategic or Co-development PartnerHome to Imme Ebert-Uphoff and Elizabeth Barnes, contributing expertise in explainable AI, machine learning for atmospheric sciences, and academic course development.
• North Carolina State UniversitycoreStrategic or Co-development PartnerHome to Ruoying He focusing on physics-inspired AI for offshore oceanography including OceanNet development for Gulf of Mexico and U.S. East Coast.
• University at Albany, SUNYcoreStrategic or Co-development PartnerHome to Christopher Thorncroft and Kara Sulia leading winter weather research including road condition classification and precipitation type prediction.
• New York State Department of Transportation (NYSDOT)coreStrategic or Co-development PartnerCollaborative development of AI model using 2,400 traffic cameras for real-time automated identification and prediction of road weather conditions statewide.
• American Meteorological SocietycoreChannel Partner/ Reseller/ DistributorOngoing partnership for conference presentations, educational programs, and teacher training initiatives. AI2ES hosts dedicated sessions at AMS Annual Meetings.
• Radiant Earth FoundationminorStrategic or Co-development PartnerCollaboration on benchmark datasets for neural network attribution methods and MLHub data distribution for earth science applications.

Scale indicators8 records

Recent moves6 records

Expansion highlights12 records

Direct peers

• National Center for Atmospheric Research (NCAR): AI2ES's closest collaborator, NCAR is a federally funded research center focused on atmospheric and Earth system science. They co-develop the CREDIT digital intelligence twin and share researchers (Gagne, Demuth, Schreck, Cains), making them peers in AI-for-weather research and operations.
• Radiant Earth Foundation: Active AI2ES partner on benchmark datasets and MLHub distribution for earth science. Comparable as a non-profit focused on open geospatial and earth science data/machine learning infrastructure.
• Allen Institute for AI (AI2): Non-profit AI research institute focused on foundational AI research with strong emphasis on open science and trustworthy AI. Comparable as a similarly structured non-profit research institute pursuing high-impact AI research with open-source releases.

Emerging players

• Google DeepMind (GraphCast / Weather Lab): Developed GraphCast, a leading AI weather forecasting model. Comparable as it pursues AI-based weather prediction at scale, but with vastly larger compute and commercial deployment channels.
• NVIDIA Earth-2 / FourCastNet: NVIDIA's Earth-2 platform and FourCastNet AI weather model represent an industry-grade competitor in AI weather forecasting. Comparable as it builds AI weather models at infrastructure scale, though with a hardware-anchored go-to-market.

Broad incumbents

• UK Met Office: UK's national weather and climate service, actively developing AI-based weather forecasting methods. Comparable as a public-sector weather forecasting organization investing in AI/ML for operational prediction.
• ECMWF (European Centre for Medium-Range Weather Forecasts): Intergovernmental organization operating one of the world's premier weather forecasting centers and a leading AI weather research program. Comparable to AI2ES in pursuing AI-driven forecasting but at a much larger scale with global intergovernmental backing.
• The Weather Company (IBM): AI2ES's strategic partner for transitioning research models to operations, particularly for cold-stunning events. Comparable as a commercial weather AI deployment organization, though operating at much larger commercial scale.
• NASA Goddard Institute for Space Studies: NASA's climate and atmospheric research center with parallel AI/ML application to Earth system modeling. Comparable as a public-sector research organization pursuing AI for weather and climate science.
• NOAA (National Oceanic and Atmospheric Administration): AI2ES's core government partner for research-to-operations. Comparable as the primary US federal weather and ocean science agency that itself funds and develops AI weather and climate tools.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat6 records

Key risks7 records

Key highlights7 records

Customer concentration

• GitHubgithub.com/ai2es
• LinkedInlinkedin.com/company/ai2es

Revenue estimate

Valuation estimate

Number of profiles

Profiles12 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about NSF AI Institute on Trustworthy AI for Weather, Climate, and Coastal Oceanography (AI2ES)