Know Your Agent: The Book That Names the Thing Nobody in Banking Has Named Yet

As AI agents gain authority inside financial institutions, a new book argues the industry needs a governance discipline built specifically for them.

The model is the seed. The institution is the soil it lands in. The seed grows whatever the soil can grow.”

— Stephen Bishop, Know Your Agent, Chapter 1

NEW YORK, NY, UNITED STATES, June 3, 2026 /EINPresswire.com/ — A new book released for pre order in May argues that artificial intelligence has already become the most consequential unmanaged relationship in financial services and that the industry has, so far, been addressing it with the wrong vocabulary, the wrong frameworks, and a borrowed confidence that is going to run out.

Know Your Agent, written by Stephen Bishop and Tony del Fierro, makes a deceptively simple argument: AI agents are a new kind of party. Not a vendor. Not a model. Not software in the legacy sense. A party. It accumulates trust through use, acquires practical authority inside live workflows, and operates deep inside the regulated relationships between financial institutions and their customers. The existing disciplines KYC, KYB, AML, vendor management, model risk, were not built for this. And the gap between what institutions think they have approved and what is actually happening inside their workflows is widening by the day. Know Your Agent The Incident That Opened the Book Bishop and del Fierro open with a community bank 8-K filing from 12 May 2026, what appears to be the first US financial institution to self-report an AI related customer data exposure at scale.

An employee had been routing daily work through an unauthorised public AI tool. Customer information entered the tool’s interactions. The institution had a written AI policy. Employees had been briefed on it. None of it was enough. The authors point out that the institution was not hacked. No vendor was breached. The exposure was a workflow. And the word that keeps them coming back to the filing is self-report because the question it raises is not about the institution that disclosed. It is about every institution that didn’t. A Discipline, Not a Policy The book’s central contribution is the concept of Know Your Agent (KYA) as a formal governance discipline for financial institutions in the age of AI. The argument draws on the history of how banking built its other disciplines: KYC, AML, BSA.

Each started as a problem no one had named. Each became a regulatory expectation. Each eventually became a discipline that shaped how banking is done. KYA, the authors argue, is the next one.

And like its predecessors, it starts earlier than most institutions think, not with the agent, but with the institution itself. The discipline they call Know Your Ent erprise, or KYE, asks institutions to read their own policies, procedures, audit findings, exception logs and committee minutes the way they would read a new customer’s KYC file: honestly, and against actual behaviour rather than documented intent.

Why Now The timing is not coincidental. On 17 April 2026, the OCC, Federal Reserve, and FDIC issued revised interagency model risk management guidance, the first overhaul of the framework in fifteen years. The new guidance explicitly carves out generative and agentic AI, stating they are ‘novel and rapidly evolving’ and outside its scope, while promising a separate request for information ‘in the near future. ‘ In practice, banks must self-govern AI risk with no formal rulebook for the foreseeable future. Simultaneously, the infrastructure is arriving regardless of readiness.

FIS, the financial technology company powering nearly 12% of the global economy, announced a strategic collaboration with Anthropic to bring agentic AI into core banking operations, beginning with an AI agent designed to compress anti money laundering investigations from hours to minutes. Fiserv, which serves 6,000 financial institution customers including 3,000 running its core systems, launched agentOS, an agentic AI operating system, built with OpenAI and Amazon Web Services. AI is not arriving through the front door of a procurement decision. It is arriving through the infrastructure thousands of community banks and credit unions use every day, whether or not their AI committee has reached a conclusion.

Know Your Agent is available for pre-order now. kyabook.com About the Authors Stephen Bishop is a financial technology strategist, podcast host, and cofounder of multiple fintech ventures. He has spent two decades working at the intersection of banking infrastructure, regulation, and emerging technology. Tony del Fierro is a financial services executive and governance specialist with extensive experience in risk management, compliance frameworks, and institutional AI deployment.

Sources: OCC Bulletin 2026-13 (occ.treas.gov); FIS press release, 4 May 2026 (fisglobal.com); Fiserv agent OS launch, 14 May 2026 (investors.fiserv.com); Davis Polk analysis of 2026 MRM Guidance (davispolk.com)

Alfie Brown
Financial Times – NY
+1 917-555-0199
email us here

Legal Disclaimer:

EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.

Media gallery