Switzerland has established a new standard for AI governance and deployment in financial services with the introduction of a FINMA-aligned technical blueprint. This initiative, a collaboration between Swiss-based companies LatticeFlow AI and Unique AI, translates the Swiss Financial Market Supervisory Authority (FINMA) principles into specific technical assessments, providing financial institutions with audit-ready evidence to evaluate, deploy, and continuously oversee AI systems across banking-specific use cases, including Know Your Customer (KYC), chatbot assistants, and Anti-Money Laundering (AML).
The assessment, detailed in an available blueprint, was conducted by LatticeFlow AI on Unique AI’s Investment Insights Agent, an agentic AI solution developed to personalize investment recommendations and rationales for relationship managers and client advisors. This aims to facilitate faster, more informed decision-making. The blueprint maps FINMA’s guidance (FINMA guidance 08/2024) to measurable technical controls, such as testing, monitoring, explainability, and model robustness, demonstrating how banks can transition from abstract policies to verifiable, audit-ready evidence.
LatticeFlow AI’s assessment evaluates the practical behavior of an agentic AI system, providing evidence on its consistency, reliability, responsiveness to changing inputs and user interactions. Crucially, it also verifies whether human users can understand, challenge, and override the system’s recommendations when necessary. This capability is vital for financial institutions, where AI informs decisions but ultimate accountability remains with human teams.
This assessment was applied to an AI system already implemented across financial institutions. Unique AI is recognized as a rapidly growing fintech company, serving over 40 institutional clients and more than 30,000 financial professionals globally. Leading Swiss and international institutions, including Pictet, Julius Baer, BNP Paribas, SIX Group, and Helvetia, currently utilize Unique AI’s technology and can now apply this new blueprint to strengthen their AI governance with concrete evidence.
Dr. Petar Tsankov, CEO and Co-Founder of LatticeFlow AI, stated, “Trust in AI is built through concrete evidence, not through abstract policies. By connecting FINMA’s guidance and principles with deep technical assessments, we provide banks with the evidence they need to make informed decisions and accelerate AI adoption safely and responsibly.”
Dr. Sina Wulfmeyer, Chief Data Officer at Unique AI, added, “In financial services, innovation only scales when it is built on trust. As AI becomes part of core investment and advisory workflows, banks need continuous technical evidence that these systems are reliable, transparent, and safe to use. This blueprint reflects our commitment to building AI that meets the expectations of Switzerland’s highly regulated financial sector and can be deployed with confidence in practice.”
Beyond this specific initiative, the blueprint underscores Switzerland’s increasing influence as a global benchmark for practical, evidence-based AI governance. By translating regulatory expectations into thorough technical assessments applicable in real financial environments, this blueprint developed by LatticeFlow AI and Unique AI demonstrates a distinctly Swiss approach to fostering global trust in AI.
