Defensible scope
VERIOK-AI certifies present facts: AI usage that is declared, limited and verifiable. Each certificate precisely defines its scope and exclusions so that a third party understands exactly what is covered and what is not.
When required, the certificate can be reinforced through eIDAS-aligned timestamping issued by accredited external trust service providers.
We certify a formal, verifiable declaration of AI usage: where AI is used, for which operational purpose, in which processes, and under which explicit limits.
We certify the declared AI environment: tools, providers and, where applicable, relevant models or configurations.
We certify the existence of defined human controls: review, approval, intervention and escalation. We do not assess intentions, only verifiable controls.
We certify that proportional minimum evidence has been submitted and referenced: internal documentation, links, excerpts, configurations or examples. Evidence requirements are proportional to risk and context.
Each certificate includes a public ID and an integrity hash allowing verification that the document has not been altered since issuance.
In practice, VERIOK-AI certification is used to reduce friction and support informed decisions.
We define scope, request proportional minimum evidence and make it verifiable without turning it into a heavy audit.
