WhaleSpotter thermal AI starts whale alerts in San Francisco Bay

AI system

System: FLIR thermal cameras with WhaleSpotter AI detection, human verification and Whale Safe alert mapping
Model: Not assessed
Release: Not assessed
Application developer: WhaleSpotter (detection AI) with UC Santa Barbara's Benioff Ocean Science Laboratory
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Thermal cameras observed an area used by whales and vessels. People reviewed each AI detection before Vessel Traffic Service operators relayed alerts by radio. [1] [2] [3]
Human oversight: Not assessed
Relation to AI: AI detects candidate whale signatures in thermal imagery. Human specialists verify them before alerts reach mariners. IEEE Spectrum adds an early detection count and the developer’s explanation of why human review remains necessary. Bay Nature independently observed the Coast Guard workflow. Neither report independently estimates the system’s effect on whale survival. [1] [2] [3]
AI contribution: Not assessed

Outcome identifiers: Not assessed
Benefit: WhaleSpotter detected whales in thermal images, and human reviewers passed alerts to mariners. Independent reporters documented Coast Guard use but no measured decline in strikes. [1] [2] [3]
Category: Improved Environmental Sustainability
Status: Not assessed
Beneficiary groups: Society. Organizations. People. [1] [2] [3]
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Participants: Developer: WhaleSpotter (detection AI) with UC Santa Barbara's Benioff Ocean Science Laboratory. Operator: Benioff Ocean Science Laboratory with the U.S. Coast Guard Vessel Traffic Service and The Marine Mammal Center.
Jurisdictions: Benefit: San Francisco, United States. Developer: United States. Operator: United States.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Conditions for replication: Whales continue to enter the bay in the detection footprint. Mariners and ferry operators act on posted detections. Funding sustains camera nodes and specialist review.
Replication attempts: 1 follow-up entries. Benefit recurred: 1.
Later findings: 2 July 2026. Bay Nature reported that the ferry camera had been installed and that Vessel Traffic Service operators were using Whale Safe. No comparative strike outcome was reported. [1] [2] [3]
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Source numbers: AIBD-0252-source-0001–0003 [1] [2] [3]
Citations: Numbered references below. [1] [2] [3]
Language: Not assessed
Source relationships: Operator, Independent [1] [2] [3]
Original evidence: Independent reporting documents installation and Coast Guard use. It gives no measured reduction in whale strikes. An earlier infrared-camera study was identified but inaccessible, so its results are not attributed to this installation. [1] [2] [3]
Access and versions: The UC Santa Barbara report is available as both a webpage and a PDF. The separate IEEE Spectrum and Bay Nature articles are also available. [1] [2] [3]
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Assessed benefits: Society. Organizations. People. [1] [2] [3]
Depth: Not assessed
Reasons: The sources document whale detections and alerts to mariners. They give no measured reduction in strikes or documented collision that an alert prevented, so they do not establish the depth of the intended benefit. [1] [2] [3]
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: Independently reported

Kind: Not assessed
Related entries: Not assessed
Field: Marine conservation / shipping. Endangered-species vessel-strike mitigation.
Purpose: The network was built to detect whales in the bay around the clock and alert vessels so they can slow or alter course.
Dates: Source published: 19 May 2026. Updated: 24 October 2026. [1] [2] [3]
Stage: Not assessed
Version: SHA-256: 0a0a7b9f5b7b…

Press release Deployer

Sonia Fernandez, “New AI system uses cameras and thermal sensors to steer ships clear of gray whales in the San Francisco Bay,” UC Santa Barbara, May 19, 2026.

UCSB is the original institutional publisher and a project participant, not an independent evaluator.
News report Independent author

Katherine Bourzac, “Thermal Cameras and AI Help Ships Steer Clear of Gray Whales,” IEEE Spectrum, May 26, 2026; reporting at the May 19 launch and interviews with the project team.

IEEE Spectrum’s reporter attended the launch and interviewed participants. The 6,600 count is a project-reported detection count, not a count of unique animals or prevented strikes.
News report Independent author

Tanvi Dutta Gupta, “Saving the Whales Just Got a Little Easier,” Bay Nature, July 2, 2026; June visit to the Coast Guard’s Vessel Traffic Service and interviews with operators and an outside whale-protection specialist.

Bay Nature adds a separate reporting visit to Vessel Traffic Service, later deployment details, and an outside specialist’s caution that captains must act on alerts.