EU AI Act risk classification and compliance report generation. Powers Irvo — from gap analysis to PDF export.
Users describe their AI system through a guided questionnaire. Claude classifies the system's risk level under the EU AI Act (minimal, limited, high, unacceptable) based on Articles 6-7. A gap analysis compares the system's current state against the Act's requirements for that risk level. Claude generates a detailed compliance report with specific actions needed. The report exports as a branded PDF with section references to the legislation.
Live as Irvo (irvo.co.uk) since March 2026. 40+ risk assessments completed. Classification accuracy validated against 15 known-risk-level systems — 100% agreement on risk tier. Average report generation time: 45 seconds. The gap analysis identifies an average of 7 compliance gaps per high-risk system. Stripe handles billing for premium reports with detailed action plans.
Irvo is BRVO's own product — built from zero to production in a two-week sprint. It demonstrates the full depth of what BRVO ships: auth, billing, AI reasoning, document generation, and deployment. Every technique used in Irvo is available to BRVO clients.
A bank deploying a credit scoring AI. Irvo classifies it as high-risk, identifies 9 compliance gaps (bias testing, human oversight, transparency), and generates an action plan with specific implementation steps.
A telemedicine platform using AI for symptom triage. Irvo flags it as high-risk under Annex III, identifies missing requirements (clinical validation, data governance), and maps a compliance roadmap.
A recruitment tool using AI to screen CVs. Irvo classifies it as high-risk (employment decisions), identifies transparency and bias testing gaps, and generates documentation templates for compliance.