Built on patented quantum measurement science, Q-Verity doesn't just check if a case exists — it verifies whether the AI told the truth about what the court held, and it does so in real time.
In June 2023, attorneys Steven Schwartz and Peter LoDuca were sanctioned $5,000 by Judge Kevin Castel in Mata v. Avianca for submitting a brief with six fabricated case citations generated by ChatGPT. The cases sounded real. The citations looked real. None of them existed. Since then, the problem has accelerated — and evolved. Today, the most dangerous hallucinations aren't missing cases. They're real cases with fabricated holdings.
The question is no longer whether you need to verify AI-generated citations. It's how — and how deeply.
Checking if a citation exists is straightforward — look it up in a database. But what about when the case is real and the AI lies about what it held? That's the harder problem, and it's where Q-Verity's patented science operates.
Surface-level checks. The case is real, so it passes — even when the AI fabricated the holding.
Measures the geometric relationship between the AI's claim and the court's actual opinion using a patented quantum measurement technique.
Any tool can check if a case exists. Only Q-Verity measures the alignment between what the AI claims a case held and what the court actually held — detecting subtle misstatements like "applied strict scrutiny" when the court actually applied rational basis review, or "upheld the statute" when it was struck down. This capability is patented and cannot be replicated by database-lookup approaches.
Q-Verity works where you work — inside your browser while drafting with AI, and inside Microsoft Word for final review. Two environments, one verification engine, complete coverage.
Write your brief, memo, or motion in ChatGPT or any AI assistant. Change nothing about your workflow.
One button, right in the AI interface. Q-Verity scans every citation in the response.
Every citation checked for existence against authoritative case law databases and holding accuracy across 144 dimensions.
Each citation flagged as verified, mismatched, not found, or contradicted — with an explanation of why.
Q-Verity sits in the Word ribbon. Run a full-document citation audit before filing.
Produce the verification record courts require — formatted for attachment to filings under AI disclosure standing orders.
Q-Verity is not built on prompt engineering or statistical heuristics. It is built on a patented measurement apparatus that operates at the hardware level during inference, analyzing the geometric structure of truth across 144 dimensions.
Every citation verified against authoritative case law databases. If a citation isn't in the record, it doesn't exist.
Exceeds prior state-of-the-art on standard public hallucination-detection benchmarks under 5-fold cross-validation.
Each citation measured across 144 interpretable dimensions of a neural cross-encoder — from surface token similarity to deep structural alignment. Every flag comes with an explanation of where the divergence occurs.
Correct references produce signal. Random references produce none. This is not a statistical artifact — it measures real structure in the geometry of meaning.
International patent protection
The dual-boundary controller for maintaining informational equilibrium during inference is protected by international patent filings. The quantum measurement technique that powers Q-Verity's holding verification cannot be replicated by classical approaches.
Developed by EISM
Built by the European Institute of Science in Management. Validated on 29,887 items across four benchmarks under rigorous 5-fold cross-validation.
Inference-time verification
Q-Verity's patented controller operates during inference — analyzing the AI's output at the moment of generation, not after the fact. A stateless electronic governor that modulates execution without retraining the underlying model.
We ran Q-Verity against the actual brief filed by attorneys Schwartz and LoDuca — the filing that resulted in $5,000 in sanctions and made international headlines. The results:
Fabricated citations caught. Every phantom case identified as nonexistent in the record.
Additional holding misrepresentations detected in citations to real cases that a database-only tool would have passed.
Total verification time for the entire brief. A task that took the court weeks to uncover.
A database lookup catches the six fake cases. Q-Verity catches those and the three holdings that were subtly wrong — the kind of error that survives a basic existence check and misleads the court.
Q-Verity serves every part of the legal system that relies on accurate citations.
Protect your attorneys from sanctions. Verify every AI-generated citation before it leaves the firm — in the browser during drafting, in Word before filing.
Learn more →Give clerks the tool to verify the citations in every filing. Enforce AI disclosure standing orders with scientific verification, not self-certification.
Learn more →Ensure that AI-assisted legal analysis across agencies meets the highest standard of citation accuracy and holding verification.
Learn more →All plans include a 14-day free trial. No credit card required.
$99/month is less than 6 minutes of billable time at standard rates. One sanctions motion costs $5,000 minimum — plus the reputational damage that no amount of money repairs. One malpractice claim costs your career. Q-Verity is not an expense. It is the cheapest insurance a litigator can buy.
See full plan details on the pricing page.
The sanctions in Mata v. Avianca were $5,000. The reputational damage was incalculable. Don't be the next headline.
No credit card required. Browser extension + Word add-in. Working in under two minutes.