Legal data, sourced and verified

Effective, verified AI tools for the practice of law.

Public Vector is a company focused on sourcing, verifying, and processing legal data. We use public and proprietary data-mining techniques and sources to create and support legal technologies. We offer deep research tools for attorneys, compliance checks for IT teams and in-house counsel, employment referrals, self-service tools for non-attorneys, and more. Each of these practices reflects a single overarching mission: to strategically source and combine legal expertise and resources into effective, verified AI tools that let attorneys expand their practices, reduce overhead, and confidently provide more efficient and effective legal services to their clients.

Open to read, updated every day

No account needed — settlement benchmarks, the searchable database behind them, and a free preview of each morning's briefing.

Always watching

Vigilance never sleeps: three monitoring layers sweep the consumer economy on a continuous cycle.

Web surveillance

Agents that patrol the consumer web itself.

Autonomous browser agents continuously sweep consumer-facing sites — reading marketing claims at scale, exercising consent banners, walking pricing flows, and recording the trackers that fire — so violations are observed as they happen on the defendant's own pages, not reconstructed after the fact.

Live-site agents · claim scouts · consent & pricing probes
Complaint surveillance

The consumer's own voice, monitored at scale.

We continuously monitor where consumers already report harm — regulator complaint databases, consumer-protection boards, forums, and review sites. A violation only advances when independent complaint volume corroborates it, so evidence arrives with putative class members already on the record.

CFPB narratives · BBB reports · Reddit · review boards
Signal surveillance

Science, trends, and the docket — watched daily.

Literature monitoring flags products whose claims new research contradicts. Daily grievance time-series spot accelerating harm categories before they make news. And every candidate clears a prior-litigation screen — federal docket plus the open web — so what surfaces is first-mover.

PubMed / OpenAlex · complaint trend series · CourtListener
Sectors & violation types under watch
Wiretap & session recording (CIPA) Cookie-consent defects GPC / Do-Not-Sell ignored Biometric capture (BIPA) Data-broker identity sync False "all natural" labeling Made-in-USA origin claims Greenwashing FDA-misclassified health claims Deceptive pricing & fake sales Junk fees & drip pricing Subscription dark patterns Web accessibility (ADA) Browsewrap enforceability
Dashboard

Screen the defendants yourself

Every defendant in the dashboard carries an arbitrage gap score between zero and one — the distance between what the site does and what its policies actually disclose. Filter, sort, drill in, and pull a sealed PDF on any defendant you want to pursue.

  • 100,000+ pre-audited defendants — consumer websites and mobile apps
  • Filter by industry, gap score, tracker category, ToS provisions
  • Per-defendant detail: third-party recipients, PII flags, consent UX
  • One-click sealed PDF with FRE 901 chain of custody
Marketplace

Buy a sealed, exclusive case packet

Each marketplace listing is a defendant our agent has already verified live: verbatim representations captured, network traffic recorded, wrap classification done, sealed PDF assembled. The first buyer claims the listing exclusively — once sold, it's gone.

  • Pre-vetted defendants organized by case lane
  • Three tiers: Standard $250 · Premium $500 · Flagship $1,000
  • Exclusive on sale — no two firms get the same packet
  • Drop-in evidence: complaint, opposition brief, class-cert record

How a defendant becomes evidence

Surveillance spots the violation. This pipeline turns it into evidence.

Step 1

Crawl & capture

Our agent drives the defendant's site like a real user — submits forms, hits checkout, exercises consent banners — while a network listener records every tracker that fires and a representation extractor pulls verbatim marketing language.

Step 2

Score the gap

We cross-reference the captured tracking behavior against the defendant's own privacy policy, ToS, and cookie disclosures. The result is an arbitrage score: the documented distance between what the site does and what it tells consumers.

Step 3

Seal the record

Every artifact — screenshots, HAR captures, policy snapshots, case-law alignment — is hashed and bundled into a sealed PDF with FRE 901 chain of custody. Drop it straight into a complaint or class-cert backup record.

Where our data comes from. Everything in a report is first-party: our own agents drive the defendant's live public site and capture the network requests, marketing representations, and consent UX exactly as they served — nothing licensed, nothing second-hand, every artifact hashed at capture. Scope details in the FAQ.

New here? The full walkthrough covers the daily briefing, the case marketplace, the records database, and a sample sealed PDF.

— The Public Vector team