Machine Learning Assurance Whitepaper

Finding Trust and Transparency with ML


Machine Learning presents new challenges for governance, risk, compliance, and audit professionals. Some of the top areas of concern are:

  • Maintaining consumer privacy 
  • Ensuring compliance with current regulations
  • Preparing for future AI regulations
  • Following ethical practices for responsible AI

For these reasons, and more, we have to be able to understand, inspect, and test the decisions made by ML systems.

Machine Learning Assurance (MLA), is a controls-based process for ML systems that establishes confidence and verifiability through software and human oversight.

Achieving Machine Learning Compliance

Download our whitepaper for a deeper introduction to Machine Learning Assurance and an overview of how Monitaur's software can support your assurance efforts.

In this 10-15 minute read, you’ll gain insights into:

  • How regulatory developments are driving the need for Machine Learning Assurance
  • Principles and best practices with MLA
  • How to use CRISP-DM as a framework for MLA
  • How MLA can be deployed in industry-specific use cases

Key Definitions - Getting Started with Machine Learning Assurance

Machine learning

Machine Learning (ML) is a form of artificial intelligence whereby computer systems recognize patterns and make predictions or decisions without explicit programming.


Machine learning models are the mechanisms needed for a machine to recognize patterns and learn how to make decisions. A machine learning model is comprised of training data, algorithms, and other important information.

Machine learning assurance (MLA)

MLA is a controls-based process for ML systems that establishes confidence and verifiability through software and human oversight.

AI regulations

AI regulations are laws that define how organizations use and report on AI systems, especially when it comes to consumer privacy, ethical practices, and transparency.


Cross-Industry Standard Process for Data Mining is a framework for how ML is conducted by practitioners. CRISP-DM was specifically tailored to enhance machine learning assurance in 2018.

The high-level steps of CRISP-DM methodology include:

  • Business Understanding
  • Data Understanding
  • Data Preparation
  • Modeling
  • Evaluation
  • Deployment 

Download our whitepaper to learn more about our expanded methodology that includes the top 10 controls for machine learning.


Getting Started with MLA

Thank you! Your submission has been received. Please check your inbox for your copy of the ML Assurance whitepaper!
Oops! Something went wrong while submitting the form.