The Hard Truths of Enterprise AI Adoption | Jodi Blomberg (Cox Automotive)

**Luz** (0:08)
Data Storytellers, we're here again. I think now this is an annual kind of habit, an annual custom that we have Jodi Blomberg on the show during the summer. We had one, I think, in 2023, we had one in 2024, and now it's 2025 And we're catching Jodi in a very exciting stage of both her career, but I think out there, the data and analytics and AI landscape, some big shifts happened over the past years. And there's a lot to talk about. So Jodi, first of all, welcome back on the show.

**Jodi Blomberg** (0:44)
Thank you for having me. You know, I'm just as nice in the winter, Luz.

**Luz** (0:49)
We haven't tried it yet, but I know that our summer episodes are always great. So someone who likes what works. But jokes aside, so what changed since 2023 and 2024? I remember the previous episode that we were talking about was about GNI expectations. That's like 2024 summer, right? Just kind of seeing, okay, that's great that we have so much energy going into this. And very cool that the business is open to innovation and experimentation with these new technologies. But just, you know, some mild warnings, I think, in that episode, more like, you know, let's just be smart about this and make, let's make good investments. And now in 2025, again, a lot, a lot happened, and now everyone is talking about AI strategies in their businesses, and deployments show different results. Kind of people are looking for the structures, the frameworks, the playbooks, you know, what's actually, what's actually good that leaders should follow. It's a little bit, again, it's a little bit the Wild West. So how do you see the landscape today?

**Jodi Blomberg** (1:58)
I mean, agents have exploded the landscape, I feel like, right? Everyone's really excited about those. And we're being what our CPO calls pragmatically aggressive. So we launched an AI accelerator in January, and pretty quickly we're inundated with these cases.
And a bunch of them, frankly, weren't agentic AI. And some weren't even LLMs, some were just ML. But neither of you are there, we were there to accelerate. So we have put out, we have pilots right now, we have three agentic solutions actually in pilot, and we're for internal use cases, so we can more aggressively go after our externalized cases and learn our lessons. And what we mean about pragmatically aggressive is, there's just a lot of unknowns on how to deploy an agentic solution, how to build one cost effectively from latency, cost, controlling for hallucinations. So we're trying to build up some scaffolding, so we have some, we're not just total no man's land, but the only way to turn those unknowns into knowns is to go do it. So we're doing it.

**Luz** (3:16)
And I just had a great conversation with Elena from TE Connectivity. She's the chief data officer there. And I don't know if you know Elena. She's been around a lot. She was like the group chief data officer at the known. And then before Nestle and Johnson and Johnson. And back in like early 2000s, she was leading the digital transformation of the Wall Street Journal. Right. So that's and she kind of said that, interestingly enough, this whole AI revolution, AI transformation is more like the digital transformation that the data transformation narrative that we had.
Yeah, right. So it's very interesting because I remember when we were in the data analytics, data science is the second professional of the 21st century stage. I remember Gartner came up with the four stage analytics maturity model. Right. So I'm now kind of looking at the landscape for the new ones. Okay, what will be our AI maturity model? Right. So we can kind of see that, okay, maybe the pinnacle of that in this era will be day agentic solutions actually deployed at scale in the business, like generating concrete value across key areas. I know it's kind of early to call, but now you are actually actively thinking through AI strategy. And as you think through an AI strategy, obviously, it's good to have some maturity scale. But people, I find, I see that people are just branding it themselves because, you know, you actually need to look at what you're doing in the business and to come up with that framework. So how do you see this AI maturity model today? Have you thought about it in that way?

**Jodi Blomberg** (5:00)
We have thought about our maturity. I think of it in two ways. One, any company today needs to get some AI maturity and just how they work as a company, right? Like, are your software engineers being more productive? Are you handing out copilot to everybody, right? There's this transformation that happens internally. And then there's a transformation in what products you go to market in or how you revolutionize your business, right? Not just your day-to-day knowledge workers, but are you putting AI into your product that you sell or that you use? And we're sort of running at both of those. One of those is a whole company transformation and it cannot be said enough how much change management that is. And so I think that's a level of maturity. And I think we talk a lot about how that's kind of table sticks now. Right? Like you can make your salespeople a lot more effective because they can pull like a really nice conversation piece on someone they're meeting with, but broken over other salesperson. It's not really differentiator as table sticks. So you don't want to be behind, but you got to do it. On the business side and the go-to-market side, you're really then you're trying to think about how to differentiate, how to provide experiences for your customers, that they can get nowhere else or take advantage of data only that you have. We're thinking a lot about what that means, right? What that data differentiation means and what to hold on to and what to give out. You already see a pretty wild change, I think, in sites run on ad revenue, for example, right? I think The Economist said 7 out of 10 Google searches. Nobody's clicking on a link anymore. That's really changed the whole SEO game too, right? And so, I think about maturity, like I said, I think about the internal maturity, like an organization embracing AI from top to bottom. You don't need to live in a product org or an AI org to do that. And then really, what are you going to market with? Are you revolutionizing your business? Are you understanding? Are you mature enough to understand what got into you, where your opportunities lie, and do you have enough scaffolding to deliver it too, right? Like, this feels a lot to me, like the early days before we had ML Ops, when we had to come up with the new paradigms for releasing this thing that doesn't look quite like the other things we do, right? Agentic AI and LMs are part machine learning, part software engineering, and we don't quite have frameworks yet, I think, on how to scale those effectively and safely, manage risk, monitor over time, observability, evaluation. And I think that's where, until we get a little bit of that scaffolding, it's going to be really hard to scale up.