Your buyers stopped taking your word for it.

Deals stall in security review because self-reported evals stopped counting.

The AI section asks for model provenance, hallucination controls, and evidence that outputs get evaluated. An internal dashboard doesn't answer that, because your own team produced it.

One measurement, reused across every deal for a year.

Two tiers, kept separate. Where your stack exposes token probabilities, the instrument-grade tier reads the model's own internal signals. Its published characterization is 0.852 AUC (fabricated-entity discrimination, length-controlled, leave-one-model-out) in the lab's method paper (DOI: 10.5281/zenodo.21365654). Where it doesn't, a documented behavioral protocol measures your fabrication rate against known answers and reports its own error rates from your engagement rather than borrowing the instrument's number.

Reviewers will probe one thing in particular: the measurement can't be coached. Context injection changes what a model says, not the weight-level geometry the score reads, and those experiments are printed in the method paper with their boundary conditions. The way to raise a score is to train the knowledge in.

Two questions your buyers ask that now have measured answers.

Training-Data Ingestion Screen, $8,500 per training event ($4,500 alongside an Evidence Pack): a screening measurement of whether your fine-tune absorbed a specified corpus, with positive and negative controls. It turns the "we never train on your data" paragraph into evidence. Its limits travel with it: meaningful-content ingestion rather than rote strings, and screening language, never a certification.

CHAI Applied Model Card supplement, $2,500: measured results formatted into the model-card fields health-system buyers now request.

The Evidence Refresh, $5,000 per quarter keeps it current, and founding vendors lock $4,000 per quarter for two years. Full numbers on the pricing page.

The founding cohort: three engagements, by selection.

Vendor engagements in the founding cohort run at $6,500, invoiced only on delivery, in exchange for permission to describe the engagement in a short case study you approve in writing. Named is preferred, anonymized is accepted at the same rate. The invoice carries a written guarantee: if the pack gives your sales file nothing you can put in front of a buyer's reviewer, say in writing what is missing within fourteen days and the invoice is cancelled. The rate is low on purpose, because the quarterly refresh relationship is the real business. Selection favors vendors whose buyers are already asking for independent evidence. Three founding engagements exist in total, shared with the line for organizations that run AI. After them, the price is $12,500 and it stays there.

The five questions vendors ask first