DeepMind's GDMI beats NHC Atlantic track forecasts at 12–72 h

AI system

System: Google DeepMind AI hurricane forecast model, ensemble mean designated GDMI in NHC guidance
Model: Not assessed
Release: Not assessed
Application developer: DeepMind
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Real-time availability of the Google DeepMind ensemble to NHC for most of the 2025 season. NHC's standard homogeneous verification procedure applied to early-model versions of the guidance.
Human oversight: Not assessed
Relation to AI: NHC's own homogeneous verification isolates GDMI's track and intensity performance against official forecasts, consensus aids and other models. The sample excludes early-season storms, when GDMI was unavailable.
AI contribution: NHC's own homogeneous verification isolates GDMI's track and intensity performance against official forecasts, consensus aids and other models. The sample excludes early-season storms, when GDMI was unavailable.

Benefits

Outcome identifiers: Not assessed
Benefit: beat official Atlantic track forecasts at 12–72 h
Category: Improved Task Accuracy
Status: Not assessed
Beneficiary groups: People. Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Measurement

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: beat official Atlantic track forecasts at 12–72 h
Comparison: the NHC official track forecast (OFCL) in the 2025 Atlantic homogeneous model comparison
Results: Reference: OFCL errors of 20.4, 31.1, 41.9, 49.2, 62.2 and 81.4 n mi at 12, 24, 36, 48, 60 and 72 h. Result: GDMI errors of 19.6, 26.4, 34.8, 41.4, 50.8 and 66.5 n mi at the same lead times.
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Money

Kind of value: Not assessed
Amount: Not assessed
Currency and prices: Not assessed
Period and unit: Not assessed
Realization: Not assessed
Recipients and payers: Not assessed
Gross or net: Not assessed
Included costs: Not assessed
Comparison: Not assessed
Method: Not assessed
Return and payback: Not assessed

People and places

Participants: Developer: DeepMind. Operator: NOAA National Hurricane Center.
Jurisdictions: Benefit: Miami, United States. Developer: United Kingdom. Operator: United States.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Follow-up

Conditions for replication: Consistent timely delivery of the AI guidance (NHC notes some AI systems were not consistently available). Results from one season. NHC's Science Operations Officer says it takes a long time to fully evaluate new models.
Replication attempts: 2 follow-up entries. Benefit recurred: 2.
Later findings: Nothing after the 30 March 2026 report was found in the cited sources.
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Evidence

Source numbers: AIBD-0089-source-0001–0004
Citations: Numbered references below.
Language: Not assessed
Source relationships: Operator
Original evidence: Not assessed
Access and versions: Not assessed
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Grades

Assessed benefits: People. Organizations. Society.
Depth: Not assessed
Reasons: People: Aggregate verification statistics, no discrete event. Organizations: Aggregate verification statistics, no discrete event. Society: Aggregate verification statistics, no discrete event.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: The deployer's account

Entry details

Kind

Not assessed

Related entries

National Weather Service storm-surge model underestimated Hurricane Sandy's threat and NJ Transit lost rail equipment; NHC's hurricane forecast guidance serves the same operational decisions.

Field

Weather forecasting / tropical cyclones. Public-sector meteorological service.

Purpose

Experimental AI tropical cyclone forecast model providing ensemble track and intensity guidance to NHC forecasters.

Dates

Date basis unassessed: 2026 (precision unassessed). Updated: 22 October 2026.

Stage

Not assessed

Version

SHA-256: 3d3a6d9a0d41…

References

Other identifiers: AI Incident Database: 536.
Category mappings: Not assessed
Brief: Not assessed
Revisions: 8 revisions. Latest: 22 October 2026.

Sources

Report Deployer

Cangialosi, J. P., Martinez, J., Reinhart, B., Papin, P. (2026). National Hurricane Center Forecast Verification Report: 2025 Hurricane Season, Section 1 (Introduction), p. 7. NOAA/NHC, 30 March 2026.
Press release Deployer

National Weather Service (n.d.). AI in Hurricane Forecasting at the National Hurricane Center: A Q&A with NHC Science Operation Officer Wallace Hogsett. NOAA/NWS News Around NOAA.