FOR LITIGATION PRACTICE

The filing was drafted by someone else's AI. Verify every citation before you build your response on it.

Briefs, motions, demand letters, and discovery responses now arrive drafted with AI, and the errors read exactly like the correct material around them. A fabricated case, a superseded regulation, or a holding quoted to say something it does not say is invisible on a read, including to the lawyer who filed it.

In a 2026 California appellate matter, a fabricated citation traveled from a Reddit post into a sworn declaration, into an attorney's filings, into opposing counsel's proposed order, and under a judge's signature, because every professional in the chain assumed someone upstream had checked. The court sanctioned the introducing attorney $5,000 and referred the matter to the State Bar. A public database now tracks more than 1,000 court and tribunal decisions worldwide addressing fabricated AI content in filings. Asking another AI to check the document is not a fix. In our pre-registered study of twenty company profiles, three frontier checkers disagreed by as much as 9.6 to 1 on the same document, and the harshest one returned zero errors on documents containing sixteen verifiable ones.

The AI is the witness. The math is the judge. GauntletScore verifies each claim against primary sources, then computes the trust score with deterministic math. The agents gather the evidence; no model sits in the verdict. The same evidence always produces the same score.

What GauntletScore checks
Case citations against court records, so a cited authority resolves to a real decision that holds what the document asserts.
Statutory and regulatory references against current regulatory text, so the document does not rest on a provision that has changed or been invented.
Named entities, dates, and events against the public record, so the parties and the history in the document are real.
The argument's causal structure because a filing is a chain of reasoning, and a chain that does not hold is the response you write.

GauntletScore runs a dedicated analysis of cause-and-effect claims, testing temporal order, proportionality, and logical structure. An argument whose causal chain fails the test counts heavily against the document. The engine is built to catch the filing that reads persuasively and does not hold together.

What the first run looks like

Upload the document at gauntletscore.com. Minutes later you are reading the Gauntlet Report: a trust score with a 95% interval, every flagged claim with its verdict and the primary source behind it, and a cryptographically signed, tamper-evident certificate. In a live production run, the engine examined nineteen claims in one fluent, credible document and returned two debunked against the court record, each with the source that contradicts it. Every voting agent independently recommended against proceeding.

LIMITS

It does not make your argument for you and it does not verify what no public source can confirm. A claim it cannot check is reported as unverifiable, not as false.

Where it fits
Incoming filings.

Run opposing counsel's brief the day it arrives, and know which of its authorities fail before you respond to them.

Pre-filing review.

Verify every citation, reference, date, and calculation in your own draft before it is signed and filed.

Inherited files.

Test the checkable specifics of a matter you took over before the strategy rests on them.

There is a filing on your desk right now whose citations nobody on your side has tested. Run it. Either they hold and you know the fight you are in, or one fails, and that failure is your response.

Analyze a Document
Three free credits. No card. No call.

Validation study pre-registered on the Open Science Framework, March 2026; in progress, manuscript in preparation.