Internet-scale traffic classification corpus

Name: Internet-scale traffic classification corpus: labelled decisions on every request seen across the advertising and application estates — human vs bot vs AI-agent identification, and for agentic commerce a record of each AI agent's action sequence from product discovery through checkout (target routes, actions taken, allow/block verdicts, inferred intent), plus account/checkout/business-logic abuse signals and an annual AI Traffic & Cyberthreat benchmark.
Creator: HUMAN (HUMAN Security)

HUMAN (HUMAN SECURITY)A5 · direct accessManage supplier listing

Internet-scale traffic classification corpus: labelled decisions on every request seen across the advertising and application estates — human vs bot vs AI-agent identification, and for agentic commerce a record of each AI agent's action sequence from product discovery through checkout (target routes, actions taken, allow/block verdicts, inferred intent), plus account/checkout/business-logic abuse signals and an annual AI Traffic & Cyberthreat benchmark.

Open buyer room →

Sample

Brickroad customers

Sample this dataset before you buy it.

Your sourcing agent asks HUMAN (HUMAN Security) and files the sample in your catalog — private to you.

Start 7-day trialNothing is charged until the trial ends.

Coverage

Universe, instruments and categories.

Industry: Application Software
Instruments: equities
Categories: AlternativeSentimentReference

Dataset card

Type: not stated
Format: Tabular request/decision event logs (request-level verdicts, agent action sequences, route and outcome fields) with Text metadata (agent identity, intent classification, policy reason codes)
Volume: Internet-scale: the advertising line alone implies trillions of impressions/requests annually across participating platforms and publishers, with a labelled verdict per request; the agentic-commerce subset is a small but fast-growing slice of that stream
Users: N/A — per-request measurement platform rather than a registered-user product; the meaningful population is the measured internet audience, which for an ad-verification estate of this size is in the hundreds of millions of unique devices monthly
History: 10 years
Update frequency: not stated
Growth: Active — new AgenticTrust / Sightline / Agentic Visibility modules, a Q2 2026 Forrester Wave leadership placement, Fall 2026 G2 grid leadership, and a 2026 AI Traffic & Cyberthreat Benchmark report
Launched: ~2012 for the core traffic-classification platform (company operating as HUMAN, formerly White Ops, founded in the mid-2010s and rebranded 2021); the agentic-commerce module is ~2025-2026
Delivery: not stated
Entity mapping: not stated
Sample: not stated
Point-in-time: not stated
Licence: not stated

supplier index read from their site