Datasets
Buyer room
SHADOWFLEET.AIManage supplier listing
Shadow Fleet AI — Investigated dark-vessel event corpus with evidence trails: a global AIS fusion stream (global AIS + coastal receivers + port-call data, ingested at up to ~19M positions/day, positions refreshing as often as every 6 minutes) run through detection models that flag dark activity, transponder spoofing, identity swaps and ship-to-ship rendezvous events — then, uniquely, closed out by AI agents that investigate each event against ownership, beneficial-owner/shell-company layers, port-call history and pattern data to return a CONFIRMED identification with a plain-language reason a…
supplier index read from their site
Investigated dark-vessel event corpus with evidence trails: a global AIS fusion stream (global AIS + coastal receivers + port-call data, ingested at up to ~19M positions/day, positions refreshing as often as every 6 minutes) run through detection models that flag dark activity, transponder spoofing, identity swaps and ship-to-ship rendezvous events — then, uniquely, closed out by AI agents that investigate each event against ownership, beneficial-owner/shell-company layers, port-call history and pattern data to return a CONFIRMED identification with a plain-language reason and a traceable paper trail, in a median 11 minutes. Scale disclosed on-site: ~10,000 AIS-gap events detected daily, 62,000+ commercial vessels observed weekly, 61 flag states in the designated fleet with Panama and Russia together accounting for almost half. Latent asset: a daily stream of adjudicated (not merely flagged) maritime evasion events, each with an explanation string and an evidence chain — a labelled, agent-verified ground truth most AIS vendors never produce because they stop at the alert. Worked example published on the site: a tanker on Ukraine's war-tracking list went dark at 06:14 UTC on 19 Aug 2026 from a last position at 37N 001E and did not transmit again for 14.6 days.
Scale LargeFrom 3 yearsCoverage Industrials
Sample, licence terms, pricing and eval results when Shadow Fleet AI publishes them. Until then, discover alternatives today with a 7-day trial.
Start 7-day trial →What buyers ask about Shadow Fleet AI, answered from this page.
Shadow Fleet AI — Investigated dark-vessel event corpus with evidence trails: a global AIS fusion stream (global AIS + coastal receivers + port-call data, ingested at up to ~19M positions/day, positions refreshing as often as every 6 minutes) run through detection models that flag dark activity, transponder spoofing, identity swaps and ship-to-ship rendezvous events — then, uniquely, closed out by AI agents that investigate each event against ownership, beneficial-owner/shell-company layers, port-call history and pattern data to return a CONFIRMED identification with a plain-language reason a…
Shadow Fleet AI offers (Alternative, Geospatial, Supply Chain, Sentiment) — Large — stated ingest is 'up to 19M positions a day', i.e. roughly 7 billion position records per year; at a conservative ~50-100 bytes per fused position that is on the order of 0.4-0.7 TB/year of raw stream, before derived layers. Derived layers: ~10,000 classified gap events/day (about 3.6M/year) each with confidence, reason string, evidence set and a confirmed-identity resolution; plus a 62,000-vessel weekly universe across 61 flag states with ownership/shell-company graphs and port-call history. The adjudicated event log with its explanations is the part worth buying; the raw position stream is expensive to store and largely replicable from AIS resellers..
Sanctions compliance and enforcement screening for refiners, traders, charterers, insurers and ports, where the confirmed identification plus paper trail is what satisfies an audit; defence and maritime domain awareness for agencies tracking grey-zone logistics; training and evaluation data for AI systems — the daily stream of investigated events (evidence in, confirmed identity plus evasion verdict plus written rationale out) is the supervised-and-explained pair needed to train or evaluate agentic investigation models, and the 62k-vessel / 61-flag-state graph is a beneficial-ownership resolution benchmark; physical-crate and commodity alpha, since sustained dark activity and rendezvous clusters front-run observable supply of sanctioned crude; maritime insurance underwriting on dark-fleet exposure within a covered fleet.
The data is with 3 years of history.
Coverage spans US, UK, Other; Defense & Space; alternative, geospatial, supply_chain, sentiment; equities, commodities.
Shadow Fleet AI has not added their own details yet. Not yet on file: