These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
FAIAS (Financial AI Independent Assurance Standard)
FAIAS (Financial AI Independent Assurance Standard) is a baseline for reviewing generative, agentic and predictive AI systems in financial institutions. The author describes the current version, 1.3.3, as emerging thinking rather than finished work, and invites contributions. FAIAS supports design review, pre-production testing, independent validation, internal audit and model risk review.
The baseline is built around 109 Independent Review Questions. Each one is a precisely worded question that reviewers can answer through evidence, safe non-production testing and interviews. The questions cover nine domains: data, model, agentic behaviour, human oversight, security, third parties, operational resilience, conduct and fairness, and governance. Each question has a review canvas and a step-by-step test procedure. Tests are rated from one to four stars for how established their method is. No test is rated five stars, because no settled financial-services benchmark exists yet.
The questions support 28 assurance claims that must hold before a system is deployed. Systems are sorted into four tiers, so a lending system receives far deeper review than a simple assistant. Evidence is graded by confidence level, from a bare assertion to operational evidence. Reviewers can conclude with an independent assurance opinion: proceed, proceed with conditions, do not proceed, or unable to form an opinion. FAIAS draws on registers of regulatory sources, academic papers and documented AI incidents. It is released under a CC BY 4.0 licence.
About the tool
You can click on the links to see the associated tools
Tool type(s):
Objective(s):
Target sector(s):
Country/Territory of origin:
Lifecycle stage(s):
Type of approach:
Maturity:
Usage rights:
Stakeholder group:
Risk management stage(s):
Technology platforms:
Tags:
- ai governance
- model risk management
- llm security
- ai assurance
- agentic ai
- audit
- financial services
Use Cases
Would you like to submit a use case for this tool?
If you have used this tool, we would love to know more about your experience.
Add use case
Partnership on AI
