Your AIMS needs evidence about the model. Not just about the process.
ISO/IEC 42001 certification reviews your AI management system: policies, roles, processes. The clause that quietly demands more is performance evaluation, because process documents cannot answer how the model behaves. Spectralgraph is an independent AI-testing laboratory that supplies that missing artifact.
Certification bodies audit the management system. Nobody measures the model.
The ISO/IEC 42001 wave is real: accredited certification bodies are actively certifying AI management systems, and organizations are assembling evidence for performance evaluation, monitoring, and continual improvement clauses. Management-system certification confirms your governance operates. It does not, and structurally cannot, measure your model. That is the gap this measurement fills, from outside your organization, with no stake in the outcome.
It's not a checklist or a vendor questionnaire. What it does is measure the model itself, reading the model's own internal signals rather than grading its response output. Using your real question types on a model you host, it maps where the model fabricates, how often, and on which topics, down to specific flagged answers your team can verify with their own eyes. The deliverable is a signed, dated screening measurement built for your governance file. In testing so far it catches fabricated answers at about 0.85 AUC on models it has never seen before; that figure is provisional until the work is published.
No customer data. Nothing installed. Five business days.
The test needs representative question types and a model endpoint you designate; it never touches customer or user data. In scope are models you host or control, including private Azure OpenAI and AWS Bedrock deployments and fine-tunes. The first measurement gives your AIMS a dated baseline; quarterly re-measurement produces the monitoring trend record the standard's lifecycle clauses point at.
The same artifact serves NIST AI RMF programs: it is MEASURE-function evidence, produced independently.
Fixed fees, written engagement, nothing success-based.
Fees are fixed and never success-based: the LLM Validation Report is $9,500, Continuous Assurance is $6,500 per quarter for one model, and the pre-deployment Model Selection Study runs $4,500–$7,500 by candidate count. Full details and the founding client program are on the home page. Every engagement runs under a written agreement before any work begins.
Questions AIMS owners ask first
Does this certify us against ISO/IEC 42001?
Where does the measurement fit in an AIMS?
What does the test need access to?
Our certification body already reviews our AI governance. Why add measurement?
Put a measurement behind the management system.
Reply and we will send the sample validation report so you can see the exact artifact your evidence file would hold.
Email the lab