"Because the algorithm said so” isn’t good enough anymore.

When your machine learning model makes a decision that affects someone’s medical treatment, financial security, or legal rights, stakeholders need to understand why.

I first encountered interpretable machine learning working in insurance - though I didn’t realise it at the time.

The insurer I worked for used machine learning models as part of its premium calculation process. There was an unwritten rule that any models we deployed had to be easily explainable to our policyholders.

My employer was a government work cover insurer and the motivation behind this rule was simple. For people to have confidence in the work cover system, the calculation process needed to be transparent.

As a result, model interpretability was implicit in everything my team did.

That was more than a decade ago, and the stakes have only gotten higher.

And while AI and machine learning seem to be increasingly heading in the black-box direction, with deep learning and LLMs, the need for interpretability has never been more critical – particularly in high-stakes fields like medicine, law and government, where algorithmic decisions can fundamentally alter the course of someone’s life.

Serg Masis has literally written the book on machine learning interpretability, and given his book, Interpretable Machine Learning with Python, is now in its second edition, the topic clearly resonates strongly with many data scientists.

In the latest episode of Value Driven Data Science, Serg joins me to share practical strategies for building interpretable machine learning models that earn stakeholder trust and accelerate AI adoption within your organisation.

You’ll learn:

1. The crucial distinction between interpretable and explainable models [07:06]
2. Why feature engineering matters more than algorithm choice [14:56]
3. How to use models to improve your data quality [17:59]
4. The underrated technique that builds stakeholder trust [21:20]

Model accuracy means nothing if stakeholders don’t trust the predictions.

Listen now on Apple Podcasts or Spotify, or click the link below:

​Episode 98: Building Trust in AI Through Model Interpretability​

Talk again soon,

Dr Genevieve Hayes