Regulatory Compliance

Global compliance, built in.

AegisOS Comply exports audit trails in regulator-accepted formats for financial institutions operating across 8+ jurisdictions. Every AI spend decision is documented to the exact standard your regulator expects.

8+
Regulatory frameworks
6
Countries / regions
1-click
Compliance export
100%
Audit trail coverage
🇮🇳
India
South Asia
✓ RBI Compliant
✓ SEBI Aligned
✓ IT Act Compatible
Export format
RBI JSON / CSV
RBI Guidelines on AI/ML
Reserve Bank of India · Effective 2024

The Reserve Bank of India mandates that financial entities using AI or ML models for customer-facing or automated financial decisions maintain explainable, auditable records of every model output and the rationale behind it.

Full audit trail of every AI-driven spend decision with timestamps and policy match
Human-in-the-loop escalation pathway for high-value or high-risk transactions
Model explainability — which rule fired, at what priority, on what inputs
Minimum 7-year retention of financial transaction audit records
Export in structured format suitable for RBI inspection
Audit trailExplainability7-year retentionHuman oversight
SEBI Circular on Algorithmic Systems
Securities and Exchange Board of India

SEBI requires broker-dealers and financial intermediaries using algorithmic or AI-driven systems for order placement or financial decisions to maintain detailed logs of every system-generated action.

System-generated flag on every AI-initiated transaction
Risk parameter documentation and override audit log
End-to-end traceability from agent decision to payment execution
Algo auditSystem flagRisk parameters

🇪🇺
European Union
Europe · 27 member states
✓ EU AI Act
✓ GDPR Compatible
✓ PSD2 Aligned
Export format
EU AI Act JSON / CSV
EU Artificial Intelligence Act
European Parliament · In force August 2024

The EU AI Act classifies AI systems used in credit scoring, insurance risk assessment, and financial decisioning as high-risk, triggering mandatory requirements under Articles 9–15 and Article 13 (transparency).

Article 13: Human-readable explanation of every AI decision made about a natural person
Article 9: Risk management system with logged risk scores per transaction
Article 12: Automatic event logging with timestamps, inputs, and outputs
Article 14: Human oversight mechanism — approval workflows with documented responses
Article 17: Quality management system records retained for 10 years post-deployment
High-risk AIArt. 13 transparencyArt. 14 oversight10-year retention
GDPR — Data Processing Records
General Data Protection Regulation · Article 30

GDPR Article 30 requires data controllers and processors to maintain records of processing activities involving personal data. AI-driven financial decisioning that processes personal data falls under this obligation.

Records of processing activities with data categories, purpose, and retention
Automated decision-making disclosure under Article 22 — right to human review
Data minimisation — only necessary data attributes logged per intent
Art. 30 recordsArt. 22 rightsData minimisation

🇺🇸
United States
North America
✓ SOX Ready
✓ SOC 2 Type II
✓ SEC Aligned
Export formats
SOX JSON / CSV
Sarbanes-Oxley Act (SOX)
Sections 302 & 404 · Public Companies

SOX Sections 302 and 404 require public companies to maintain and evaluate internal controls over financial reporting. AI-driven spend decisions that affect financial statements must be traceable, controlled, and auditable.

Control reference on every automated financial transaction
System-generated vs. human-approved distinction clearly logged
Approval status and approver identity recorded with timestamp
Audit trail ID for cross-referencing with financial statement line items
Immutable logs — retroactive modification cryptographically detectable
S. 302/404ICFRControl referenceApprover log
SOC 2 Type II
AICPA Trust Service Criteria

SOC 2 evaluates controls across Security, Availability, Processing Integrity, Confidentiality, and Privacy. AI spend governance platforms must demonstrate that automated processing is complete, accurate, and authorised.

Processing Integrity: every intent evaluated, decision recorded, no dropped transactions
Availability: policy engine SLA documented and measured
Security: cryptographic chain integrity, API key hashing, no plaintext secrets
Change management: policy version history with author and timestamp
Trust criteriaProcessing integrityAvailability SLA
SEC AI Governance Guidance
Securities and Exchange Commission · 2024

The SEC has issued guidance requiring investment advisers and broker-dealers to document and disclose conflicts of interest arising from AI-driven decisioning, and to maintain records of all AI-assisted investment or spend decisions.

Documentation of AI model inputs and outputs for SEC exam readiness
Conflict-of-interest audit trail for AI-recommended spend decisions
3-year retention of AI decision records under Rule 17a-4
SEC Rule 17a-4AI disclosure3-year retention

🇬🇧
United Kingdom
Europe
✓ FCA Compliant
✓ SMCR Ready
✓ ICO Aligned
Export format
Generic JSON / CSV
FCA Consumer Duty
Financial Conduct Authority · In force July 2023

The FCA Consumer Duty requires firms to deliver good outcomes for retail customers, including ensuring AI-driven systems do not cause foreseeable harm. Firms must monitor, evidence, and report on AI outcomes.

Outcome monitoring: documented record of every AI decision and its result
Foreseeable harm prevention: policy engine blocks high-risk spend categories
Board MI: analytics dashboard provides aggregate spend and denial statistics
Complaints traceability: every disputed AI decision traceable in the audit log
Consumer outcomesHarm preventionBoard MI
Senior Managers & Certification Regime (SMCR)
FCA / PRA · Accountability for AI outcomes

Under SMCR, a named Senior Manager must own accountability for AI-driven systems that affect regulated activities. AegisOS Comply's role-based approval trails support clear accountability mapping.

Named approver logged for every human-reviewed transaction
Role-based access control maps to SMCR accountability structure
Escalation audit trail — who approved, when, and what rationale was provided
Named approverAccountability mapEscalation log

🇸🇬
Singapore
Southeast Asia
✓ MAS TRM
✓ MAS FEAT
✓ PDPA Aligned
Export format
Generic JSON / CSV
MAS Technology Risk Management (TRM)
Monetary Authority of Singapore · 2021 Guidelines

The MAS TRM Guidelines require financial institutions to establish robust governance and controls over technology systems, including AI models used in financial decisioning.

AI model risk documentation — inputs, decision logic, and outcome records
Change management: policy version history with approval workflow
Incident logging: denied or escalated intents logged with full context
Cyber resilience: immutable audit chain resistant to insider tampering
Model riskChange managementCyber resilience
MAS FEAT Principles
Fairness, Ethics, Accountability & Transparency in AI

The MAS FEAT principles provide voluntary but widely adopted standards for responsible use of AI in financial services, focusing on fairness, ethical use, accountability, and transparency of AI-driven decisions.

Transparency: every decision explained by the matched policy rule
Accountability: named approver on all human-reviewed decisions
Ethical use: policy simulator tests for disproportionate denial patterns
FairnessTransparencyAccountability

🇦🇺
Australia
Oceania
✓ APRA CPS 234
✓ ASIC Aligned
✓ Privacy Act
Export format
Generic JSON / CSV
APRA CPS 234 — Information Security
Australian Prudential Regulation Authority

CPS 234 requires APRA-regulated entities (banks, insurers, superannuation funds) to maintain information security capabilities proportionate to the threats they face, with robust logging of all automated financial system actions.

Information asset classification: AI spend decisions flagged by risk level
Incident detection: anomalous spend patterns trigger risk engine alerts
Audit log integrity: cryptographic chain ensures log records cannot be altered
Third-party management: agent keypair signing verifies the source of each intent
CPS 234Info securityIncident detection

🇦🇪
UAE
Middle East
✓ CBUAE Ready
✓ ADGM Aligned
✓ DIFC Compatible
Export format
Generic JSON / CSV
CBUAE AI Governance Framework
Central Bank of the UAE

The CBUAE requires licensed financial institutions to establish governance frameworks for AI systems, ensuring human accountability, explainability, and auditability of AI-driven financial decisions.

AI decision explainability — which policy triggered the outcome and why
Human oversight mandate — high-risk transactions require named approver
Risk classification per transaction — low / medium / high / critical
Audit trail available for CBUAE inspection on demand
AI governanceHuman oversightExplainability
ADGM & DIFC Regulatory Frameworks
Abu Dhabi Global Market & Dubai International Financial Centre

Firms operating within the ADGM and DIFC free zones are subject to their own financial services regulators (FSRA and DFSA respectively), both of which require documented governance of AI and automated financial systems.

Conduct risk records for AI-driven client-affecting decisions
System controls documentation for regulatory examination
Conflict-of-interest audit trail for automated spend decisions
FSRADFSAConduct risk

🇨🇦
Canada
North America
✓ OSFI B-13
✓ PIPEDA Aligned
Export format
Generic JSON / CSV
OSFI Guideline B-13 — Technology & Cyber Risk
Office of the Superintendent of Financial Institutions

OSFI B-13 sets expectations for federally regulated financial institutions in Canada regarding the governance, risk management, and controls over technology systems — including AI and algorithmic decisioning.

Technology risk inventory: AI spend systems classified and documented
Change and configuration management: policy version history with timestamps
Cyber event logging: all denied and escalated intents logged with context
Third-party dependency controls: agent signing verifies external AI integrations
OSFI B-13Tech riskChange management

Need a framework that's not listed?

Our compliance team can assess your jurisdiction's requirements and configure AegisOS Comply's export format to meet your regulator's exact specifications. Enterprise plans include custom format support.

Talk to our compliance team →