Statement of Guidance

Guidance on the use of AI in Jersey's financial services sector

Jersey Financial Services Commission (JFSC) · Jersey

Status not confirmed

Published: 0001-01-01

Current version last checked: 2026-07-26

Summary

This is non-binding guidance issued by the Jersey Financial Services Commission (JFSC) explaining how financial services firms can use Artificial Intelligence (AI) responsibly. It expressly does not create new legal or regulatory requirements; instead it explains how firms should apply their existing obligations (e.g. under AML/CFT, data protection and cybersecurity regimes) in a risk-based and proportionate way when deploying AI.

  • Governance and accountability: Boards and senior managers should maintain clear ownership, oversight and accountability for AI-assisted processes, including human oversight of higher-impact decisions and stronger controls for autonomous 'agentic AI'.
  • Cybersecurity, privacy and data quality: Firms should integrate AI into existing cybersecurity and data protection frameworks, ensure lawful bases for processing personal data, validate input data quality, and manage third-party AI dependencies under existing outsourcing policy.
  • Regulatory compliance: Firms remain responsible for AI used in compliance functions, customer communications and regulatory reporting, and must be able to verify accuracy and maintain appropriate human review.
  • Consumer protection and fair treatment: Firms should test AI outputs for bias, avoid harm to customers, and ensure complaints and challenge processes cover AI-assisted decisions.
  • Transparency and explainability: Firms should disclose AI use to customers where relevant and maintain an AI register recording use cases, data inputs, oversight and risk classification, supported by illustrative scenarios and a materiality ladder (low, medium, higher impact) to guide proportionate governance.

The guidance closely aligns with OECD AI Principles and is intended to be proportionate: low-impact productivity tools need only light-touch controls, while higher-impact uses (onboarding, lending, investment advice, fraud detection) warrant stronger governance, testing, monitoring and record-keeping. Firms remain accountable for AI-driven outcomes just as they would for any other technology or process.

Applies to

financial services firms regulated in Jersey

Topics

Version history

2026-07-26

source file (current)