AI system
- System
-
Google Flood Hub
- Model
-
Not assessed
- Release
-
Not assessed
- Application developer
-
Google - Role
-
Not assessed
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
Global AI flood forecasts available for ungauged Malian rivers. Evaluation against discharge observations and text-mined district-level impact data across lead times and thresholds.
- Human oversight
-
Not assessed
- Relation to AI
-
The study directly evaluates Google Flood Hub’s machine-learning forecasts. It measures forecast performance, not a humanitarian outcome or a deployed change to the Mali Red Cross’s early-action protocol.
- AI contribution
-
The study directly evaluates Google Flood Hub’s machine-learning forecasts. It measures forecast performance, not a humanitarian outcome or a deployed change to the Mali Red Cross’s early-action protocol.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
skillful beyond 4-day lead time, enough for 7-day Mali Red Cross early action triggers
- 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
-
skillful beyond 4-day lead time, enough for 7-day Mali Red Cross early action triggers
- Comparison
-
Not assessed
- Results
-
skillful beyond 4-day lead time, enough for 7-day Mali Red Cross early action triggers
- 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:
Google. Operator: Evaluated for the Mali Red Cross Early Action Protocol; not documented as an operational trigger. - Jurisdictions
-
Developer:
United States. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
Forecast thresholds and matching windows require local validation; performance differs substantially by district. Frequent false alarms need to be considered alongside the potential for earlier warnings. The proposed change to early-action planning requires implementation and outcome evaluation.
- Replication attempts
-
No attempts recorded.
- Later findings
-
The study proposes changes to early-action planning; implementation and subsequent benefits are not established.
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0092-source-0001
- Citations
-
Numbered references below. [1]
- Language
-
Not assessed
- Source relationships
-
Independent [1]
- Original evidence
-
Some observations are confidential, and the impact data are available only on request. The EarthArXiv preprint and conference presentation were inaccessible. [1]
- 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: Skill evaluation, no discrete event. Organizations: Skill evaluation, no discrete event. Society: Skill evaluation, no discrete event.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
Independently studied
Entry details
- Kind
-
Not assessed
- Related entries
-
Google-developed hazard early-warning system failed to alert millions; Flood Hub is Google's flood equivalent, with the same flood-exposed public as beneficiaries.
- Field
-
Hydrology / flood forecasting. Humanitarian anticipatory action.
- Purpose
-
AI-based riverine flood forecasting for ungauged basins to support early warning and anticipatory action.
- Dates
-
Source published: 14 March 2025. Updated: 24 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: 6a7e928484f2…
References
- Other identifiers
-
AI Incident Database: 1546.
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
8 revisions. Latest: 24 October 2026.
Sources
-
Peer-reviewed study Independent author
Kuipers, E. R., Oldenburg, V., Phung, P., Sutanudjaja, E., Ficchi, A., Salamon, P., van den Homberg, M. (2026). Assessing the Riverine Flood Forecast Skill of GloFAS and Google Flood Hub With Impact Data and Discharge Observations to Support Early Actions in Mali. Journal of Flood Risk Management, doi:10.1111/jfr3.70247.
independent