Data-Focused AI Adoption Workshop Topics

Your team of leaders want AI “fully functional” in your organization right now. ROI from day one! If you’ve been in tech a while, this harried, demanding stance is not new. The money being thrown at it isn’t, either. Just more of it. The adoption advice to slow down and go faster is also not new. And yet it’s what’s needed for a solid start at your AI adoption. Let’s look at three big problems: legacy systems need rewiring, most company data is unusable, and there are compliance/legal risks. I’ve made a very short list of considerations for each topic. A few lists, designed to get you thinking about where to begin with your adoption workshops.

The time you spend now will save money and messes later. We didn’t want to scale crappy code; now, we don’t want to send a hurricane on top of our crumbly, crunchy house of data.

1. Legacy systems need rewiring. It’s been limping along for a while. Maybe a long while. And it’s a security-performance-data-problem. Adding AI to this won’t fix anything. Where to begin?

   1. You’ll first need to ask executive-level questions:

      1. What capabilities do we want this platform to support in x years?

      2. What are the benefits of modernization? Cost benefits? Security? The AI adoption connection? Where is our customer in this?

      3. How much change can our stakeholders lead and absorb?

      4. When we do this work, where are the key decision points or validations? This is part of your feedback loop.

2. The timeboxed data cleanup. Review data sets with the perceived highest value. Plan for long-term maintenance by building automated cleaning pipelines.

   1. Identify who can own and check this work

   2. Make a timeboxed plan

   3. Define the feedback loop with owners to check your work

   4. Once you’re through the heavy lifting, what does maintenance look like long-term, so you don’t let the mess build up again?

3. Build governance ahead of AI deployment.

   1. Create AI use policies and boundaries for all departments.

   2. Create monitoring systems for AI touchpoints.

   3. You’ll need executive-level questions here, too. Topics include:

      1. Strategy and business value

      2. Workforce and culture

      3. Human oversight and boundaries

      4. Vendor and tools

      5. Security and privacy

      6. Voice of the customer (you know, the reason you’re even having this problem to begin with?)