AI system
- System
-
Skylight maritime-monitoring platform (Speed Range Event rule and machine-learning Fishing Event detection)
- Model
-
Not assessed
- Release
-
Exact release not established.
- Application developer
-
Allen Institute for AI (Ai2), Skylight
- Role
-
Not assessed
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
A vessel transmitting AIS inside a monitored area of interest. A speed-range rule tuned to the local legal definition of trawling speed. A machine-learning fishing-behavior model producing a second alert within 40 minutes. PNA analysts integrating Skylight with the Sistema Guardacostas and a coastguard vessel within reach.
- Human oversight
-
Coast guard analysts assessed the alerts and decided to redirect the patrol vessel.
- Relation to AI
-
Skylight's own account describes a two-stage AI contribution. A rule-based Speed Range Event flagged the vessel at under 6 knots inside an area of interest. 40 minutes later a Fishing Event from its machine-learning model, which detects longlining, trawling and squid jigging, reinforced the suspicion. PNA analysts then redirected a coastguard vessel. A retired PNA captain is quoted saying Skylight contributed directly to the decision. This is the maker's account, with the deployer's officers quoted inside it. The independent report (Agenda Malvinas, citing Pescare and official information) attributes the detection to the PNA's Sistema Guardacostas and does not name Skylight or AI. So the attribution to the AI output is not independently confirmed. The PNA release Skylight links to is not among the sources.
- AI contribution
-
A speed rule in Skylight first flagged the vessel. Its machine-learning model then detected a pattern consistent with fishing. According to Skylight, coast guard analysts used those alerts when redirecting a patrol vessel.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
PNA intercepted XINRUN 579 fishing illegally in Argentina's EEZ on 5 January 2025; Skylight says its speed-range and ML fishing alerts prompted the response and a first electronic fine was paid.
- Category
-
Improved Environmental Sustainability
- 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
-
Not assessed
- Comparison
-
Assumed alternative: PNA's capability before Skylight's real-time fishing detection
- Results
-
Assumed comparison. Result: a Speed Range Event alert, then a Fishing Event alert 40 minutes later.
- Additional value from AI
-
Not assessed
- Limitations
-
The attribution to Skylight comes from its developer’s account, which quotes coast guard officers. Independent reporting names the coast guard’s own system and does not establish Skylight’s contribution.
- 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: Allen Institute for AI (Ai2), Skylight. Operator: Prefectura Naval Argentina (Dirección de Tráfico Marítimo, Fluvial y Lacustre; EICEMAR).
- Jurisdictions
-
Benefit: Argentine exclusive economic zone off Caleta Olivia, Santa Cruz,
Argentina. Developer: 
United States. Operator: Argentina. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
AIS transmissions not switched off or spoofed. A patrol asset close enough to intercept before the vessel leaves the EEZ. A legal route to sanction (Law 24.922 and the electronic fine procedure). Analyst confirmation of each alert before dispatch.
- Replication attempts
-
2 follow-up entries. Outcome unresolved: 1. Benefit recurred: 1.
- Later findings
-
31 March 2025. The administrative proceeding ended with the vessel paying the fine and the authorities' operational costs, as Skylight reported on 31 March 2025 (the payment date itself is not given; NVIDIA's 3 March 2025 blog already said the vessel was later fined).
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0350-source-0001–0008
- Citations
-
Numbered references below.
- Language
-
Not assessed
- Source relationships
-
Independent, Developer
- 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
-
People: Not assessed. Organizations: 2. Society: 1.
- Reasons
-
People: No individual beneficiary is identified. Organizations: The PNA intercepted the vessel within hours and, per Skylight, issued its first international electronic fine with the fine and costs paid. The independent report confirms the interception and administrative proceedings. Society: One vessel among hundreds operating at the edge of Argentina's EEZ each season; the fishery-level effect is not quantified.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
Independently reported
Entry details
- Kind
-
Not assessed
- Related entries
-
Not assessed
- Field
-
Fisheries enforcement / illegal, unreported and unregulated fishing. National coastguard using a free non-profit AI platform.
- Purpose
-
Real-time detection of suspicious vessel behavior and fishing activity to support maritime enforcement against illegal, unreported and unregulated fishing. The interception was the intended use.
- Dates
-
Date basis unassessed: 2025 (precision unassessed). Updated: 21 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: 30e9f88a9087…
References
- Other identifiers
-
Not assessed
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
7 revisions. Latest: 21 October 2026.
Sources
-
News report Independent author
Agenda Malvinas (2025). Prefectura detectó un buque asiático pescando ilegalmente en aguas argentinas. 8 January 2025 (source line: Pescare).
la Prefectura Naval Argentina detectó este domingo 5 de enero, al buque pesquero XinRun 579, de bandera de Vanuatu, realizando pesca ilegal en la Zona Económica Exclusiva Argentina
Tras la intercepción del Guardacostas GC-25 Azopardo, se notificó formalmente al capitán del XinRun 579 sobre la infracción
-
Press release Developer
Skylight (2025). Real-Time Intelligence Powers Argentina's Fight Against Illegal Fishing. skylight.global, 31 March 2025 (maker's account).
Just 40 minutes later, a Fishing Event alert provided additional intelligence, reinforcing suspicions that the vessel could be engaged in illegal fishing activity.
then by a machine learning model that detects several types of fishing behavior like long lining, trawling, and squid jigging
Skylight contributed directly to the Coast Guard's decision-making by delivering immediate and reliable insights.
The case led to Argentina’s first international electronic fine using AI-powered evidence.
-
Press release Independent author
Wolfberg, E. (2025). AI Model Offers Conservationists New Tools to Protect Fisheries, Wildlife at Scale. NVIDIA Developer blog, 3 March 2025 (technology-partner blog, not the system's maker).
Argentina's Coast Guard interdicted—and later fined—a vessel illegally fishing in its exclusive economic zone after Skylight alerted it to the vessel's suspicious behavior
-
Press release Developer
Skylight surfaces signals. The inspections, fines, and decisions belong to the agencies that made them.