UK DUAA right to complain enforces in -- days · 19 June 2026 EU AI Act high-risk enforces in -- days · 2 August 2026

For General Counsel

The proof a regulator asks for, before they ask.

Article-mapped evidence. Decision provenance produced continuously. Contestation workflow that satisfies the right-to-challenge requirements emerging across jurisdictions. The artefacts a regulator actually wants, in the format their statutory framework requires.

If the system verifies itself, the evidence will not hold.

Not documentation. Evidence that holds.

Observer Mode shows how your system behaves

Enforce Mode blocks actions before execution

If an action violates policy, it does not run

The legal problem

Documented intent does not satisfy enforcement requirements.

Most AI governance produces declared posture: policies, attestations, periodic audits. Regulators are increasingly asking for something different: evidence that the control was active at the moment the decision was made, that the decision is traceable, and that the affected individual can challenge it.

OBEXGATE produces those artefacts as a side effect of operation. Not assembled later. Not reconstructed from logs.

What it produces

Three classes of artefact, every governed decision.

→ Decision provenance

Per-action record of what was evaluated, which frameworks applied, which rule was determinative, why the decision was made. Tamper-evident.

→ Article-mapped evidence packs

Evidence is mapped to the article, principle, or control reference inside the source regulation. Cross-mapped where a single event triggers multiple frameworks.

→ Audit lineage

Continuous, verifiable chain across every governed event. Cryptographically linked. Discoverable in regulatory inquiry.

Frameworks covered

Eighty-eight enforcement engines across eight jurisdictions.

Each engine maps to specific articles inside its source regulation. Cross-mapping handles the case where a single decision is governed by multiple frameworks simultaneously.

Jurisdiction Frameworks (selection)
European Union EU AI Act (Article 99 tiers, SME proportionality applied), GDPR, DORA, eIDAS, NIS2
United Kingdom UK GDPR, Data Protection Act 2018, UK DUAA, FCA, PRA
United States HIPAA, FDA, CCPA, NIST AI RMF, FedRAMP, CMMC, SOC 2, ISO 27001
Brazil LGPD (Article 52, BRL 50 million cap per violation)
Australia Privacy Act 1988 (s13G greater-of structure, AUD 50M floor for serious interference)
New Zealand Privacy Act 2020 (Section 212 criminal fines, Human Rights Review Tribunal damages)
Canada PIPEDA, AIDA, Ontario Bill 194
Singapore PDPA, Model AI Governance Framework

Complete article-level coverage detail.

Contestation

Right to challenge an automated decision.

The EU AI Act, GDPR, and several other frameworks require a workable mechanism for affected individuals to challenge automated decisions. OBEXGATE includes a structured contestation workflow that produces the record of the challenge, the human review, and the resolution as part of the audit lineage.

This is a regulatory requirement under several frameworks. It is also the artefact that demonstrates substantive compliance, not declared compliance, in regulatory inquiry.

Decommissioning

Structured system retirement, not deletion.

When a governed system is retired, OBEXGATE executes a sequenced teardown: credentials revoked, access removed, retained data purged where required, audit trail closed and sealed, identity record decommissioned. Every phase is recorded as part of the audit lineage.

This satisfies regulatory requirements for AI system disposal under EU AI Act lifecycle obligations, ISO 42001 lifecycle documentation, GDPR Article 17 right-to-erasure flows, and SOC 2 asset decommissioning controls.

See the evidence your governance would produce.

Six questions. Personalised regulatory map across the frameworks you are exposed to, statutory exposure detail, three-year operational governance cost basis. Or 30 minutes with the team.