They diagnose the structural design of legal procedure itself — identifying the specific rules that make bad outcomes systematically inevitable for one class of party, regardless of the underlying facts.
Legal disputes that look like they are about a specific wall, a specific expert, or a specific contract are sometimes about something else entirely: a structural gap in the design of the rules governing that dispute — a place where the legal architecture runs out, where the incentive structure goes uncorrected, and where the party who knows how to exploit the gap wins not because they are right, but because the rules make their winning structurally inevitable.
The four instruments on this platform each measure a different expression of this same underlying pattern. They produce a number — not a qualitative opinion — so the structural problem can be named, scored, and argued before litigation entrenches it.
The Competence Gap. Courts and institutions ask whether a professional is qualified to do a job. They almost never ask how close that professional's demonstrated experience is to the specific matter at hand. The gap between holding a credential and being genuinely competent for a specific technical problem is real, measurable, and currently unexamined by any formal procedure.
The Proximity Gap. Expert witnesses are required to be independent. The current test — self-declaration and adversarial challenge — has no quantitative threshold for network proximity. A direct academic or professional connection between an expert and the parties can go unchallenged simply because no formal standard for measuring it exists.
The Disclosure Gap. Most legal systems separate the rules governing disclosure before harm from the rules assigning liability after it. In the gap between those two silos, silence is free and disclosure is costly. The result is a Nash equilibrium in which the party holding private knowledge of a pre-existing problem has no incentive to share it — and the rules reward them for staying quiet.
The Escalation Gap. Standard dispute management treats case value as static. In reality, every delay, bad professional assignment, and adversarial step detonates a compounding multiplier of cost. The gap between initial claim value and actual terminal exposure is driven by interacting, rational-choice actor incentives that systematically manufacture value destruction.
Each tool takes raw inputs — case materials, practitioner profiles, expert and party records — and returns a scored diagnostic in seconds. All four tools are free, require no account, and work on any jurisdiction.
The last five years of AI in legal technology have addressed one class of problem: making existing legal processes faster. Contract review. Case research. Document summarisation. These are optimization tools — they assume the rules are correct and reduce the cost of working within them.
"A contract review engine makes a good process cheaper. A structural diagnostic engine asks whether the rules governing the contract are themselves distorted. You need the second before you can fully trust the first."
— The Procedural Gap Project · moral.moneyThese tools do not accelerate existing processes. They ask a prior question: is the structural design of this procedure producing systematically unfair outcomes? If so, for whom, under what conditions, and how severe?
| Typical AI legal tech | KYC.co diagnostic tools |
|---|---|
| Reads what is written in documents | Reads the structural design of the rules themselves |
| Makes the existing process faster | Asks whether the existing process is working correctly |
| Produces summaries, redlines, citations | Produces a score, a class, a threshold |
| Treats the rules as fixed and correct | Identifies where the rules systematically break |
| Value: cost reduction and throughput | Value: detecting structural failure before it becomes entrenched |
| Legal AI optimises the professional's process | This reads the professional's own record against their own position |
| No instrument for detecting liability migration | Detects when responsibility moves from defendant to the filing firm |
What AI enables here — that was not practically available before — is cross-domain pattern recognition. The same structural gap that appears in a Portuguese civil engineering dispute also appears in a hospital surgeon appointment, in a financial regulator's expert appointment, in an international arbitration expert panel. The pattern is the same. The gap is the same. These tools make it measurable in any of those contexts with the same instrument.
Every proposed fix to a procedural gap carries a risk: the repair can itself become a tool for bad actors. A rule requiring proximity disclosure can be gamed to delay proceedings. A rule penalising silence can be used for harassment. This is why the design of every instrument in this project obeys one constraint:
"Closing procedural loopholes that benefit bad-faith actors while ensuring the fixes are so tightly engineered that they cannot be exploited in reverse."
Each tool produces a score, not a verdict. The output is diagnostic — an evidentiary input for legal argument, not a substitute for it. A high BFM score does not mean the Passive Party is guilty. It means the incentive structure was comprehensively misaligned, and the Active Party has identifiable grounds to argue that the procedural rules were working against them. The score opens an argument. It does not close one.
The Dispute Engine does not produce a legal opinion. It produces an actuarial signal — the structural equivalent of what an aviation maintenance system does when it detects a pattern of component stress that precedes failure. The signal does not say a plane will crash. It says: this file, in this configuration, carries the structural signatures of a case that becomes something worse than it currently appears.
That reframes the natural customer. A law firm reading its own portfolio has every incentive not to find problems — every file it manages is also a record of its own conduct. A professional indemnity insurer covering that firm has the opposite incentive, and the commercial arithmetic is direct: a portfolio of a thousand live files containing three that will generate PI claims costs more to insure than a portfolio containing none. The difference between those two outcomes is currently invisible to the insurer.
The strongest signal the Dispute Engine produces is liability migration — the moment a file's structural evidence shows that responsibility is moving from the original defendant toward the filing professionals themselves. This pattern appears in the record before anyone names it, and long before a professional negligence claim is filed. It is, by definition, the signal a PI insurer most needs and currently has no systematic access to.
"The question is not whether the gaps will be found. They will be found, by whoever runs the file first."
— On Legal Bullshit: Frankfurt's Distinction and the Self-Reporting Fallacy in Litigation · Carroll (2026)Every active professional portfolio is now a candidate for this reading. The tool is free. The record exists. The structural signatures are either present or they are not. The only variable is who runs it first — the party at risk, their insurer, or the counter-party that already has access to the same file.
Note on validation: The financial exposure scores and risk signals the Dispute Engine produces are structural model outputs — not empirically validated actuarial figures. The commercial case for the instrument at scale depends on demonstrating, across historical portfolios, that the signals it identifies correlate with files that later generated professional negligence claims. That retrospective validation has not yet been conducted. What the tool identifies is structural; what it predicts remains to be tested against outcome data.