AI Governance

Most organizations are no longer deciding whether to use AI. They are deciding who is accountable when it is wrong. These articles cover the frameworks, model-risk expectations and operating choices that turn AI oversight from a policy document into something that works.

Enterprise AI · model risk · controls · responsible AI

AI governance frameworks

The standards organizations use to structure AI oversight.

Framework
NIST AI Risk Management Framework (AI RMF): A Blueprint for Trustworthy and Compliant AI
How NIST's AI RMF gives organizations a common structure for identifying and managing AI risk.
Framework
ISO 42001 Explained: The First AI Management System Standard
ISO/IEC 42001:2023 as a blueprint for auditable AI operations, and how it connects with NIST and regulatory frameworks.
Regulation
EU AI Act: Europe's Ambitious Gamble on Regulating Artificial Intelligence
Phased obligations, risk-based rules for high-risk and general-purpose AI, prohibited practices and fines.

Model risk

Long-standing model-risk expectations applied to AI and machine learning.

Executive
SR 11-7 in the AI/ML Cloud Era
Applying the SR 11-7 model-risk framework to AI/ML and cloud-native deployments, from board oversight to MLOps pipelines.
Fundamentals
A Guide to Federal Reserve SR 11-7 (Model Risk Management)
A plain-terms introduction to the Federal Reserve's guidance on managing the risks that come with using models.
Executive
Why Most Global Banks Underestimate OSFI E-23 Until It Changes Their Operating Model
Why reading E-23 as a vendor-governance exercise misses the larger operating-model shift.

Enterprise AI in practice

Where AI decisions meet data, identity and accountability.

Executive
AI at Scale in Banking: The Decision Supply Chain
Scaling banking AI as a decision supply chain, where data, identity, models, policy and human accountability have to work together.
Regulation
FERPA in the Age of AI: Data Governance, Privacy, and Compliance
FERPA through an enterprise data, AI and risk lens, where compliance becomes a capability rather than a checkbox.
Technology
Rust in AI
How a memory-safe, high-performance language is shaping inference, cloud and edge AI.