The CIO’s guide to managing enterprise AI | ZDNet

Today’s CIOs must ensure that their organizations have an “AI-ready” infrastructure capable of supporting data and applications related to AI.

The challenge is not whether to embrace AI, but how to prepare the organization to adopt AI in a way that increases business value and reduces risk. CIOs must also consider how to establish a distinct set of enterprise capabilities around AI. All this requires CIOs to establish a holistic view of their enterprise infrastructure and how it should evolve to support current and future capabilities related to AI and data.

On episode #706 of CXOTalk, prominent author and investor, Ash Fontana, explains how CIOs can develop an “AI-first” operating model as a framework for IT as a business capability, beyond being just a technology provider. Ash is author of the book: The AI-First Company: How to Compete and Win with Artificial Intelligence

Watch our entire, in-depth conversation and read the complete transcript. Here are edited comments from our conversation are below:

What is an AI-first company?

It can be an existing company that starts to put AI at the start of every conversation, at the top of their agenda in every meeting, or it can be a new company that is focused and strategic about collecting the right data, feeding it into the right systems, building products with predictive value.

That’s what an AI-first company is: a company that gets the imperative to build these systems and gets the need to focus on this from day one, so that you have the right data in the right place feeding into the right models. Rather than later trying to sprinkle AI over data that you’ve accidentally collected.

The implications are a focus on data management, data collection, and data talent or data-competent talent when you’re thinking about budgeting, where to focus your attention.

About investing in data infrastructure

When you’re really focused on collecting data, what are all the weird and wonderful ways you can collect that data? How do you manage, for example, a data labeling operation? It’s quite a new challenge. It’s not a function that’s existed in the past.

When you’re hiring people, what’s the difference between a product manager and a data product manager? What’s the difference between a software engineer and a data engineer? What’s the difference between a project manager and a data project manager? They’re different roles that you hire from different backgrounds.

There’s also a series of organizational questions to ask. How do you pick the right degree of centralization and decentralization in your organization? How do you centralize enough so that you have good data infrastructure, you have a good set of tools for people to use to build models all throughout your organization?

Also, how do you maintain a degree of decentralization, so data science and machine learning talent are out in the field understanding the prediction problems that your business has? Out in the field with the people with clipboards, checking off things on a safety checklist or inventory in a warehouse or whatnot. They understand what data is available, what people are trying to do, what people are trying to automate, and so on. That’s an organizational question.

Then there’s also, finally, metrics and measurement. How do the metrics differ for an AI-first company versus a normal software company? How do you really measure the return on investment in AI projects? How do you understand if these models are working? Then, how do you make sure they stay in check?

About the CIO role in enterprise AI

The role of a CIO changes to quite a degree when you think about capital allocation to data infrastructure, data collection, and data talent. It changes with respect to how you manage or structure the organization, and it changes with respect to your metrics and what you measure.

There’s so much nuance around the degree to which you focus on this, depending on where you’re at in your journey. Depending on whether you’re just experimenting with a few models to test, for example, can we actually make a prediction about demand for the thing we’re selling, demand for the apparel we’re selling next season? Can we make a prediction about a trend in this industry with our consumers, whether they’re going to buy this or that color next season? Can we make a prediction around this delivery time in our supply chain?

Are you at this stage, where you’re experimenting and trying to learn from the data that you have? The degree to which you invest in data collection, data infrastructure, and machine learning different models is very different than after you’ve done those experiments, you’re sure that you can make these predictions, and you want to double down.

What is “Lean AI”?

The Lean Startup was all about constraining my problem and my experiment to understand if customers want a feature of a product and to understand if the need is really there.

Lean AI constrains the experiment to test if customers want, or will they value, a prediction or a little automation.

There are a series of questions that help you figure out:

• What’s the one data set I need so the experiment does not, for example, require getting lots of data from lots of different places?
• What’s the one model I can use? Often, it’s a very simple statistical model rather than having a network of machine learning models that are all interlinked in a complicated way.
• What’s the one machine I can run it on? Just run it on someone’s laptop first rather than distributing across the entire computing infrastructure.

What’s the one output I can get that will be useful to people, whether it’s a chart, a one-page report, or a table of data for information?

This Lean AI process is a way for you to get up to speed quickly, doing at least one experiment that will help you figure out where to invest next.

CXOTalk presents in-depth conversations with people shaping our world. Thank you to my senior researcher, Sumeye Dalkilinc, for her assistance with this post.