Want To Measure Your Enterprise AI Initiatives? Start With Model Debt (Part 2 of 2)

Measure Enterprise AI Initiatives with Model Debt 2 of 2

Stu Bailey, Forbes – July 21, 2020

In part one of this discussion, I presented the basic concept of model debt as a way to measure the effectiveness of individual models and AI programs overall. In part two, I’ll go through a short example to show how model debt can be computed in practice.

Calculating model production debt and model value loss require the following inputs:

The target production days (TPDs), which is a count of the number of days that the model is intended to be in production over its full life cycle, starting from when the data science team releases it for production. The key factors in assessing TPDs include:

• The “lock-to-load” (LTL) time, which is the expected time between the data science team deeming a model “ready for deployment” and when the model is first deployed in production. The shorter the lock-to-load time, the faster the model can contribute to the business.

Read the Full Article

You might also enjoy

Introducing Enterprise Safeguards for Generative AI

ModelOp released version 3.2, which includes cutting-edge capabilities to govern and monitor Large Language Models (LLMs) and Generative AI — including internal and third-party models — helping de-risk enterprises while delivering value-generating AI at scale.

Get the Latest News in Your Inbox

Further Reading