Artificial intelligence is everywhere—it’s in the presentations we watch at conferences and it’s the top topic for many vendors on the expo floor. How can you digest all this information and figure out which promises are real and which are just talk—and how do you make the right choice when you’re overwhelmed? “The paralysis can be part of the problem,” explained Marc Fischer, head honcho for Inspire CRE and moderator of an AI panel at the 2026 BOMA International Conference & Expo.

The panel—which featured Sharon Hunt, VP and chief product officer for PAM Technology at JLL; Saejin Park, vice president of data and insights products at JLL; and Rick Ferrino, global chief data officer for JLL—introduced an easy-to-understand framework for AI adoption. “Crawl, Walk, Run” describes where different teams—and even individual people—fall on the scale of AI adoption and competency. Crawl describes a team or worker who’s cleaning up their data and running pilots to see how AI can help an organization. Walk is a level where you’re working AI into the things you do every day and finding ways for it to help you. People at the Run level are scaling up to agentic systems that work alongside your teams.

It’s important to resist the urge to skip ahead, the panel said. If you’re new to using AI, start at the Crawl stage and only move along when you’re confident with what you’re doing.

6 Tips for Incorporating AI into Workflows

The panel offered actionable advice for anyone looking to use AI to take things off their plate and improve the way they work.

1. Get your data together. AI trains on existing data, so make sure it’s learning the right things, advised Ferrino. “If it’s not ready to be used, it doesn’t matter what outcome you’re trying to drive,” Ferrino said.

“AI is not magic,” Park added. “If you don’t have data, you can’t get to some of these things that we’re talking about… Data by itself is worthless. Data in the service of solving a problem in a format that makes sense, tied to the business context—that is interesting.”

2. Work backwards. “In the world I’m working with right now, we first decided, what are the outcomes we want to drive? What are the things we think an AI agent or skill could be developed to enhance?” Ferrino explained. “You work backwards from there.” In Ferrino’s case, his team saw that lease data could be a highly valuable place to incorporate AI. The team had seven different sources of lease data in different formats. “Putting in effort to wrangle it into a common format is the first half of the battle,” Ferrino said.

3. Tell the right story. In the Crawl phase, training materials help sell the story of AI to your wider team, including people who may be skeptical about adopting it, Park said. “You have to be really clear about where AI is going to benefit and imagine what success looks like,” Park added. “Be able to articulate and tell the story. If you can’t tell the story of how AI is going to be helpful, you can’t get to the next step.”

4. Be selective about where you incorporate AI tools, especially at first. “Look at what’s slowing you down and what has multiple tools in play,” suggested Hunt. “When you look at technology vendors, look at those who are embedding AI deep in workflows so you’re not training teams to use a new tool. That introduces friction that can cause issues right out of the gate.” The best AI tools bake the AI workflows into the work you’re already doing so you don’t have to think about it, she said.

5. Build skills and agents that are reusable. When you’re ready for agentic AI in the Run stage, you’re ready to start scaling your tools to be useful to a wider audience. Human nuance also comes into play here. For example, you may have a client that wants a report similar to one you already produce, but in a different format. “When you’re starting to build out skills or agents, you build them in a way that they’re reusable and you can deploy them again and again—and build them in a way that allows you to still tweak,” Ferrino said. “You’re not rebuilding for every customer, you’re taking something that’s 80 to 90% done and adding a couple of tweaks.”

6. Learn where the fear is coming from. Your team or even leaders in your organization may be wary of AI, Hunt said. To combat those fears, you first need to understand why the fear exists. For example, if it’s security-related, demonstrate the safeguards you’ve put in place to protect proprietary information and ensure the tools you use have the right terms and conditions.

You can start incorporating AI into your workflows in your personal life and scale up to using them at work, Hunt added. Using them is the best way to learn them and find ways they can benefit you at work.

“AI is just a tool—the human brain and critical thinking never goes away,” Park said. “It’s more important than ever. You can create a lot of junk fast if you don’t think through what success looks like. What do you want to solve? Why do you want to solve it? How can AI help you be successful?”