# The April 2026 interagency model risk management guidance that replaced SR 11-7, explained

On April 17, 2026 the Federal Reserve, OCC and FDIC replaced SR 11-7 and OCC Bulletin 2011-12 with risk-based guidance expected to be most relevant to banks over $30 billion. It leaves generative and agentic AI models out of scope.

Explainer · Financial Services · Data & AI. Published November 14, 2026. Last reviewed November 14, 2026.

## At a glance

- Issued April 17, 2026 by the Federal Reserve Board (SR 26-2), the OCC (Bulletin 2026-13) and the FDIC.
- Supersedes SR 11-7 (April 4, 2011) and SR 21-8 (April 9, 2021). The OCC also rescinded Bulletin 2011-12, the Model Risk Management booklet of the Comptroller’s Handbook, Bulletin 1997-24 and Bulletin 2021-19.
- Expected to be most relevant to banking organizations with over $30 billion in total assets, and sometimes to smaller ones with significant model risk.
- A model is a complex quantitative method that applies statistical, economic or financial theories to turn input data into estimates. Simple spreadsheet arithmetic and deterministic rule-based software are excluded.
- Generative AI and agentic AI models are outside its scope. Traditional statistical models and non-generative, non-agentic AI models are inside it.
- It sets no enforceable standards, and non-compliance with it will not result in supervisory criticism.

## What was replaced

In April 2011 the OCC and the Federal Reserve issued model risk management guidance they had developed jointly, issued by the Federal Reserve as SR 11-7 and by the OCC as Bulletin 2011-12. The FDIC adopted it in June 2017 with technical conforming changes [3].

On April 17, 2026 the three agencies issued revised guidance. The Federal Reserve’s SR 26-2 says it supersedes and replaces SR 11-7 and SR 21-8, the 2021 interagency statement on model risk management for systems supporting Bank Secrecy Act and anti-money laundering compliance [1]. The OCC’s Bulletin 2026-13 rescinds Bulletin 2011-12, the Model Risk Management booklet of the Comptroller’s Handbook, Bulletin 1997-24 on credit scoring models and Bulletin 2021-19, the OCC’s issuance of that 2021 statement [3]. The agencies cite supervisory experience, industry feedback and advances in modeling as the reasons, and describe the new approach as risk-based and tailored to each bank’s model risk profile, size and complexity [1] [2].

## Who it is for

The guidance is expected to be most relevant to banking organizations with over $30 billion in total assets. The agencies say smaller banks’ models are typically covered by internal risk management suited to their size, so leaving them out is consistent with tailored supervision. A bank of $30 billion or less may still find it relevant if its models are prevalent and complex or it does business outside traditional community banking [2]. The OCC’s bulletin says the guidance applies to all community banks, subject to those limits [3].

## What counts as a model, and the AI carve-out

The guidance defines a model as a complex quantitative method, system or approach that applies statistical, economic or financial theories to process input data into quantitative estimates. It excludes simple arithmetic calculations, such as those in spreadsheets, and deterministic rule-based processes and software with no statistical, economic or financial theory behind them [2].

A footnote sets the AI boundary. Generative AI and agentic AI models are described as novel and rapidly evolving and are placed outside the guidance. The same footnote says a bank’s risk management and governance practices should guide the governance and controls for any tools, processes or systems the document does not cover, and that its principles apply to traditional statistical and quantitative models and to non-generative, non-agentic AI models [2].

In our reading, this splits a data and AI estate in two. A credit scoring model or a fraud classifier built with machine learning falls inside the guidance if it meets the definition. A large language model assistant or an agent that takes actions falls outside it, yet still needs governance under the bank’s own framework. The guidance does not say what that framework should contain.

## The risk-based approach

The guidance breaks model risk into four influences: inherent risk, exposure, purpose and use [2].

- **Inherent risk** comes from assumptions, complexity, input quality and data constraints.
- **Exposure** is how much the output matters to business decisions, and can be measured, for example by portfolio size.
- **Purpose** is qualitative. Models used to meet regulatory requirements or manage financial risk exposures are generally treated as higher risk.
- **Use** matters because a sound model can still carry high risk if it is misapplied or misused.

Purpose and exposure together determine materiality. A bank may deem some models immaterial, in which case model risk management may consist of identifying them and watching for conditions under which they could become material. Models of higher materiality warrant more comprehensive and rigorous oversight. The guidance also describes assessing risk in aggregate, across models that share assumptions, data or methods, as sound practice, and describes effective challenge: critical analysis by people with the expertise, independence and organizational standing to bring about change [2].

## Development, use, validation and monitoring

- **Development** generally begins with a clear statement of purpose. Testing, from out-of-sample and out-of-time tests to comparisons of alternative assumptions, should be as rigorous as the model’s complexity and materiality call for.
- **Use** depends on understanding a model’s limitations. Extending a model beyond its intended purpose calls for added analysis and a review of controls.
- **Validation** generally happens before first use. When an urgent business need requires earlier use, the guidance points to closer attention to limitations, telling stakeholders about them and controls such as usage limits or closer monitoring. Its quality depends on rigor, not on where the function sits in the organization chart.
- **Components of validation** are conceptual soundness, outcomes analysis and ongoing monitoring. For some models, interpretability measures or benchmarking may be more practical than evaluating theory [2].

## Governance, inventory and vendor models

The guidance says model risk management benefits from clear roles and accountability, including for conflicts of interest between development and validation. Where internal audit is involved, its role is generally to judge whether model risk management is rigorous and effective, not to repeat development or validation work. It describes a comprehensive model inventory as common industry practice and says an effective one holds enough information to understand risk at the individual and aggregate levels. Documentation supports continuity and the tracking of recommendations, responses and exceptions [2].

Vendor and third-party models, data and parameters remain in scope even when the vendor does not share code, data or methods. Sound practice includes understanding the vendor model’s conceptual soundness, design, development data and performance, monitoring whether it stays fit for purpose, and documenting and validating any customization [2].

## Guidance, not a rule

The document says it does not set enforceable standards or prescriptive requirements, and that non-compliance with it will not result in supervisory criticism. A footnote adds that supervisory action may still follow violations of law or unsafe or unsound practices that stem from poor management of model risk [2]. The Federal Reserve asked Reserve Banks to distribute the letter to the banks they supervise [1].

In our view, banks that built inventories and validation calendars around SR 11-7 now have three questions to settle: which tools meet the new guidance’s definition of a model, which of those are material, and where generative and agentic AI tools are governed now that this guidance leaves them out. This explainer describes the guidance; it is not legal advice.

Moraph’s teams help organizations with work like this. [See our solutions](https://www.moraph.com/solutions).

## Sources

1. Board of Governors of the Federal Reserve System. SR 26-2: Revised Guidance on Model Risk Management (April 17, 2026). <https://www.federalreserve.gov/supervisionreg/srletters/SR2602.htm> Checked November 14, 2026.
2. Federal Reserve Board, FDIC and OCC. Supervisory Guidance on Model Risk Management (SR 26-2 attachment, April 17, 2026). <https://www.federalreserve.gov/supervisionreg/srletters/SR2602a1.pdf> Checked November 14, 2026.
3. Office of the Comptroller of the Currency. OCC Bulletin 2026-13, Model Risk Management: Revised Guidance (April 17, 2026). <https://www.occ.gov/news-issuances/bulletins/2026/bulletin-2026-13.html> Checked November 14, 2026.

Published: November 14, 2026
Canonical URL: <https://insights.moraph.com/interagency-model-risk-management-guidance-2026-explained>
Publisher: Moraph Insights
