We use a few strictly-necessary cookies, no marketing trackers, and cookieless analytics. Read our cookie policy, or pick a preference.
The authoritative record of every AI system you operate: what it is, what governs it, and what you have done about it. The single place your organisation, its auditors, and its regulators go to see the truth of the estate, with every finding cited to the provision that mandates it.
Most organisations hold their AI governance in scattered spreadsheets and slide decks that no two teams read the same way. The Intelligence Centre replaces that with a single authoritative record: the systems you run, the law that governs each one, and the evidence of what you have done. Three components build on that record, and each cites its reasoning so a lawyer can verify it and a regulator can audit it.
The regulatory-intelligence engine. It reads a system's profile against the EU AI Act and the instruments it intersects, and returns structured, actionable guidance, with honest handling of contested edge cases. It informs expert judgement; it does not replace it.
Sensitive-data analysis that runs on schema definitions, early. It detects the full spectrum of regulated data, from direct identifiers and GDPR Article 9 special-category data through financial, credential, children's, location, health, and biometric fields, and flags the proxy and correlation risks that appear when innocuous fields combine.
The guided process that turns a described system into complete, compliant documentation. It screens for Article 5 prohibited practice first, then branches on your role, and each completed discovery enriches the graph, so the next similar system starts already half-answered.
Underneath sits the Regulatory GraphRAG, a knowledge graph of instruments, archetypes, provisions, and obligation atoms, and the per-tenant Customer Graph of your systems, roles, controls, and evidence. Because the two are joined, every output the Centre produces carries the article, recital, or clause that mandates it. Nothing is asserted that cannot be cited.
Source: SI obligation corpus 2026-07-18 · research-grade, internally audited · not counsel-attested
The Intelligence Centre defines what is true. Assurance keeps it true in production, engagement turns it into work, and capability proves the people can run it.