Included with every upload
The Arbitrator Report
Every arbitrator on your list, their disclosure parsed by AI and matched against every award in the database.
AI-powered arbitrator research · For plaintiff-side securities attorneys
FINRA gives you names. We give you the record.
Upload the arbitrator list FINRA sent you. Our AI has already read every award those arbitrators ever signed. In seconds you have every case, how often they side with claimants, what they award, and the background FINRA never publishes. Panel selection is the most consequential decision in the case. Make it with the full record, read by a machine that never skims.
Arbitrator report
Sample
Public chairperson · 67 awards read by AI
- Customer win rate
- 33.3%
- Awarded to customers
- $2.99M
- Award
- 7
- No award
- 14
- Expunge
- 6
- Dissents
- 0
- Chair vs. panelist
- 41 / 26
- Rule 12206 dismissals
- 2
Illustrative. Your report is built from the arbitrators on your list.
The database
Browse the public analysis- FINRA cases
- 24,845
- Arbitrator appearances
- 142,000
- Unique arbitrators
- 8,400
- Years of records
- 23.2
Not SQL. Not regex. AI.
We unleashed AI on every award FINRA has ever published.
24,845 award documents across two decades, produced by hundreds of panels and offices, no two formatted alike, the older ones scanned paper run through OCR. Pattern matching breaks on page two. A database query cannot read. So every document is read by Claude, Anthropic’s frontier model, the way a good paralegal would read it, if the paralegal never got tired and never skipped a page.
01
It reads the awards
Every award PDF since 2003. The AI finds the panel, the roles, the claims, the dollar amounts and the outcome inside documents whose layout changed a dozen times over twenty years.
02
It parses the disclosures
Your arbitrator list is a PDF of dense disclosure forms. The AI turns each one into structured data: employment, education, prior cases, conflicts, flags.
03
It resolves the names
Robert J. Smith, Bob Smith and R. Smith appear across thousands of awards. The AI recognises them as one arbitrator and links every case to a single profile.
04
It runs the Deep Dive
Thirty-plus public databases searched at once and written up as a dossier: favorability, bias indicators, a strike-or-keep call with reasons, and voir dire questions.
One award, as the AI reads it
Illustrative
…the Panel, cons1sting of ROBERT J. SMlTH, Esq. (Presiding Chairperson, Public), MARIA DELGADO (Public Arbitrator) and T. NAKAMURA (Non-Pub1ic Arbitrator), having considered the pleadings, testimony…
…Respondent is liable for and shall pay to Claimant compensatory damages in the sum of $412,000.00 plus interest… Claimant’s request for punitive damages is denied…
read by Claude
- Chair
- Robert J. Smith, public
- Panelists
- Maria Delgado, public; T. Nakamura, non-public
- Outcome
- Award to claimant
- Compensatory
- $412,000
- Punitive
- Denied
- Dissent
- None
- Profile
- Linked to Smith, R. (one profile, 67 cases)
OCR noise, inconsistent titles, three spellings of one name. Regex cannot do this. A model can.
How it works
One upload. The whole panel, on the record.
What used to take a paralegal the better part of a week, the AI does in the time it takes to open a PDF.
01
Upload your arbitrator list
Drop in the PDF FINRA sends you. The AI reads it and extracts every arbitrator name, no manual entry, no typos.
02
The AI builds structured profiles
Each disclosure form is read and parsed by AI into employment history, education, conflicts, and prior cases, then matched to every award the arbitrator ever signed. Win rates and award patterns are computed across the whole database.
03
Search the case history
Every FINRA case they have decided, filterable by outcome, amount, parties, and date. Chair versus panelist. How the decisions have trended.
04
Run a Deep Dive on the ones that matter
The AI searches thirty-plus public databases at once: voter registration, court records, FEC donations, property, bar history, social media. It writes the dossier and makes the strike call, with reasons.
What you get
Two reports. One for every arbitrator, one for the ones that matter.
Premium · Per arbitrator
The Arbitrator Deep Dive™
An AI-written opposition-research dossier on the person who will decide your client's case. Thirty-plus public databases, searched and read by AI, delivered in minutes.
“For less than the cost of a Venti Mocha Cookie Crumble Frappuccino, you can do a complete deep dive.”
The Arbitrator Deep Dive™
We do not stop at FINRA. We go find out who they are.
The award record tells you how an arbitrator has ruled. It does not tell you who they are. For the arbitrators that will decide your case, the Deep Dive sends the AI out onto the open web and into the public record: who they gave money to, what they own, who they sit on boards with, what they have said in public, and whether any of it should worry you.
It is the background check a private investigator would run, except it takes minutes, it cites every source, and it ends with a strike-or-keep call instead of a shrug.
What the dossier delivers
Per arbitrator
Where the AI looks
Money and politics
Courts and regulators
Career and affiliations
Public footprint
Are they Republican or Democrat? Do they own a plane? Who do they sit on a board with? Have they ever been sued? We dig. Then we tell you whether to strike.
What FINRA won't tell you
23.2yrs
of awards, and not one published statistic.
FINRA publishes award documents. That is it. They will not tell you how often an arbitrator sides with claimants, how much they typically award, or whether they have developed a pattern that should concern you. You are expected to read thousands of pages of PDFs and figure it out yourself. We put AI on it instead, and it read all of them.
Read the FINRA analysisClaimant win rates
Average award size
Expunged history
Community ratings
Excel is not a research tool
Somebody out there is still picking arbitrators off a spreadsheet they typed themselves. That is not research. That is a guess with gridlines.
You know the kind of file. It lives on a shared drive, it holds whatever one firm happened to notice, and every cell in it was typed by a person from memory. Nothing in it can be checked against the actual award documents, because a spreadsheet has no award documents behind it. It is not that it is wrong. It is that nobody can tell when it is.
A hand-built spreadsheet
Illustration
| Name | Cases | Good? | Notes |
|---|---|---|---|
| Smith, R. | 3 | ok | fine I think |
| Smith, Robert J | 1 | NO | same guy?? |
| Bob Smith | 2 | yes | ask Dave |
| Delgado | ruled against us 2019 (or 2018) | ||
| Whitcombe | 4 | YES | retired judge, nice |
| Vanterpool? | avoid | see email from Karen |
myArbitrator
Every award, every panel, every night
- Award documents read
- 24,845
- Arbitrator profiles
- 8,400
- Source
- Every FINRA award since 2003, read by AI, not typed by a person
- Name resolution
- AI recognises Robert J. Smith, Bob Smith and R. Smith as one person
- Award amounts
- Requested, awarded, and the ratio, per case
- Chair vs. panelist
- Computed, with the trend over time
- Dissents, punitives, fees
- Counted
- Expunged history
- The original award, before BrokerCheck was scrubbed
- Background
- AI Deep Dive across thirty-plus public databases, on demand
- Last updated
- Last night
A spreadsheet is a rumor with formatting. It can only be wrong in ways you cannot see, and the cell that matters is the arbitrator who decides your client’s case. myArbitrator’s AI has read every award FINRA has published since 2003, and every number on the page traces back to a document. There is no comparison. There is not even a category.
Panel selection
Are you ready for the truth about your panel?
Panel selection is one of the most consequential decisions in an arbitration. Most attorneys make it with incomplete information. Walk in with the full record, read by AI, and a dossier on the people who will decide.