Jason (00:48)
Hi Sean, how are you?

Sean Naismith (00:50)
Doing well, Jason, great to see you.

Jason (00:51)
Great seeing you as always. Welcome to the Wise in Five Thanks for joining us today. My pleasure. So before we dive into the five questions, I'd love for you to share your amazing background. I was fortunate to be part of your journey, but there's so much that you've done since. So please share it with the folks that are listening and watching today.

Sean Naismith (00:57)
Thank you for having me.

Absolutely. Well, I've been blessed to have a long career at the intersection of.

data science and AI with product management and financial services. And that is more appropriate today now than ever, I think with the rapid rise of AI and pretty much everything we do, it seems. So most recently was with TransUnion leading their global technical product management and innovation, rolling out their one true platform. Before that, I was at Innova International leading Innova Decision, a real time decisioning and AI platform for regulated industries.

And then of course, Jason and we worked together at LeapFrog online and what an awesome place to be. So more beyond that, but it's always been at that intersection, data science, AI product, and many times financial services.

Jason (01:59)
That's amazing. I am so happy for you. I am very proud that, you know, we could be a small part of your journey, but you've just done such amazing things. So congratulations to you. Never a surprise. So why don't we dive in on my first question is the question that I ask all of our guests, which is where do you find inspiration just in daily life? I know you just had a birthday for your daughter. I'm guessing she gives you inspiration.

Sean Naismith (02:11)
Thank you. Thank you.

Absolutely. Well, first and second, faith and family. But beyond that, I think it's the natural curiosity that many times we forget about. And I read a lot of books, articles, listen to lot of podcasts. And when you do that, there's that moment, that spark of, wow, I never thought of that before. That's an amazing new idea. But our lives are so busy that many times those sparks are fleeting and we forget about them. The ability to actually remember those, whether you write them down or not, but then to

Set aside time and actually go down the rabbit hole and see what's there. I think that's where many of the best ideas come from. So I just say that inner child, what you naturally find interesting and exciting as we consume all this new information and content and go down the rabbit holes when we have time.

Jason (03:13)
I love it. I love it. And I love starting with the faith and family. I always throw friends in there too, right? Those are the three F's, right? So I mean, as you mentioned, I know off the role that you had been serving, especially with TransUnion, deep in AI and analytics and data. And so we know what firms often struggle with AI adoption, right? It's now...

Sean Naismith (03:19)
Yeah, that's good point. 3F, yes.

Jason (03:37)
accelerating and I think people are at all different stages, but they're struggling with it because it's limited resources, there's uncertainty about where to start. So given your experience, what do you think the low hanging fruit is for folks that are looking at AI and whether it's decisioning or analytics or customer identity? A lot of them can't afford like a big data science team. So where do you feel like they should start their AI process?

Sean Naismith (04:02)
I think one of them.

greatest misconceptions right now is that the driver of AI value is going to be building the best model yourself. I think where it really stands is in the workflows and not in like automating the high risk workflows that you have like credit decisioning, but in focusing on decision augmentation. You know, when you think of your leaders and your managers within your organization, there's many tactical and strategic decisions that can be augmented with the use of AI.

In addition to that, you're going to start to build a muscle around automation, right? You're going to see those opportunities to further automate within the business. And so much of AI isn't around the model itself, it's around the surrounding workflow and the operating model to bring it to life and to use it, along with the data governance and the compliance that needs to support it. All of those are very important. Now, one thing I would say is when I look at opportunities for the application

of AI within a business, I like to boil it always down to four key things, right? Incremental revenue, revenue protection, cost avoidance, cost reduction, right? Many times we think about productivity. And when you think about where the hyperscalers have started with and the AI companies like Anthropic and OpenAI and others, a huge focus on having AI write really good code and improve itself. Because by doing that, you get to that exponential curve

of self-improvement. What does that mean for the rest of us? Well, what that means is these models are very good at things like software development and technology architecture and infrastructure. We're seeing that in our daily lives and in the headlines that are coming out. And it's an amazing time to be a builder because of that. On the same time, that needs the governance, right? We're seeing mistakes start to happen. So you need to have the human oversight and the judgment as

you roll these things out. I'd say the harder part for most companies is on the revenue side. How can you leverage AI to drive top line growth, not just bottom line efficiency? And depending on the industry you're in, some industries are gonna be disrupted far sooner than others. In understanding both the opportunities and the urgency as they relate to the AI's impact on the product itself,

how you serve customers, on how you go to market to really drive that top line growth. So low hanging fruit, what AI is good at today and where the risk is minimal and getting that operating model in place so you can actually automate and start to get to what we would see as an agentic future.

Jason (06:40)
Yep, that's really smart. And I agree with you that the majority of case studies that I've seen or the applications have been more about cost reduction, right? And unfortunately, we're seeing a lot of companies that are, know, reduction in force, you know, because of that too. I tend to believe they may be moving faster than the technology really can help them, right? Right.

Sean Naismith (06:49)
Yes.

Yes.

It's a real risk. It is a real

risk, yes.

Jason (07:04)
Little like the

air traffic controllers, right? It's like, let's fire all the air traffic controllers and then we realize no one's there to land the plane. So we don't want that to happen. ⁓ So against that framework, I know, I mean, at the companies that you'd mentioned, you've led these global platforms that have supported billions of dollars in revenue by unifying data and identity and analytics. So.

Sean Naismith (07:11)
That's right. That's right.

Jason (07:29)
How did you scale those globally? So we'll go to the other side, not for the smaller businesses, but for the larger businesses. How did you scale that globally? And what do you then see? Like all the insights that you got, what do you see for mid-sized businesses as they look to consolidate their tech stack?

Sean Naismith (07:44)
Well, one big lesson that I've learned is consolidation doesn't mean consolidating everything. You really want to focus on right sizing that platform layer. Where are you going to really benefit from having one of something? Some things we could tolerate having multiple of, and that's totally fine. But any time you create and say, we're going to have one of something that everyone's going to adopt, that now has a potential to be a bottleneck. And what you want to avoid is, especially in large organizations that are

trying to be nimble and launch new products quickly is you want to avoid bottlenecks. So when I think about a global platform, right, a way to scale something, what is the right size of that that's going to add to the efficiency and the productivity and the cycle time reduction in launching new products? So the key primitives when I think about in many of the industries like financial services, you really want to have one way of doing identity verification as an example. I don't want to get one

one answer for who Jason is in one construct, and then a different answer for who Jason is in another as an example. Also how we think about data management and data governance. We all heard and know that in order to have the best AI systems, you need the right data to power them. Well, if you have sprawling data that's mismanaged without ownership, that's another high risk, meaning higher reward for having common ways and technologies in order to do that. When I think about AI specific needs,

from knowledge basis and semantic layers and routing to large language models, there's absolutely a value in having a common governance layer, allowing teams to leverage those primitives in order to bring to life products, capabilities, internal tooling in a governed way. So that to me is very important. When I think about mid-size companies, most of us won't have the luxury to invest hundreds of millions of dollars into our own AI.

platforms. So I think this the takeaway is simple, right? Standardize the foundations, find the right partners and the tools to bring to use, and then start to build on top of that to make it happen.

Jason (09:48)
Yeah, I totally agree. And I think there's the term right drip, right? Data rich information, poor. And I think that's just going to grow exponentially now with so much more data being created and the access to it. You know, I, I, I believe that, that the barrier to entry for this is so low, but the barrier to success is so high. And

Sean Naismith (09:57)
Yes.

Jason (10:10)
And so I think we're going to see there's a lot of people who are using the tools who are not following the hygiene that you were calling out. Right. So now that you have access to it and it's relatively inexpensive and super fast, you're going to get inundated with a lot of data and maybe not know what to do with it.

Sean Naismith (10:28)
I agree. And you know, there's a risk to AI sprawl.

and a lack of governance. If you don't have governance and you're saying you have AI, you're not over the target yet. You need to focus on how is this going to be scalable. I also think there's a risk in trying to scale a prototype. If you have a lab, if you have teams that are quickly learning, don't mistake those prototypes with something that's ready to roll out to millions of customers to drive core systems. Those are different ways of thinking. And I think of the three horizons model from McKinsey on innovation.

You have to innovate today at your core, innovate on the periphery. And then there's those big bets, so to speak, that can change how you do business and your industry. And being intentional about how you think about those horizons and not conflating them, I think, is important. Because if you have a team that, the shiny object, as we all know, that has something really cool, that works well in a small prototype, making the mistake of trying to roll that out very quickly to a large

group of folks could materially impact the business. So that's coming from the experience of making sure those stage gates are in and thinking about the investment dollars the right way.

Jason (11:37)
Right, it's kind of starting slow to move fast. Great. So let's talk a little bit more about data and really privacy and compliance, right? So we know, to what we just talked about, that can trip up many firms as they start to scale. But based on the experience that you've had, what do you think that playbook is that midsize execs need to stay ahead? I think we teased out a little bit of it with your last

Sean Naismith (11:41)
That's right.

Jason (12:05)
answer.

Sean Naismith (12:05)
a little bit. You know, when I think about, there's a couple things here.

we need to be prepared for a rapidly evolving regulatory space, right? Privacy laws are probably the ones that are moving the fastest. And this is going to be true, not just in the United States by states and by industry, but also globally. So when you think about your company and where you operate today and where you want to operate tomorrow, you're going to have to plan for that flexibility in your AI technology and your AI stack. And that's not just going to be a team of one of AI engineers.

Your builders aren't just your engineers. Your builders are really everybody that's contributing to the value that's being delivered to customers. a mistake that's commonly made is bolting on compliance at the end of the build. That's never been true, and it's especially not true with AI. So building that muscle of getting the teams involved, understanding where the flexibility needs to be in the system in order to adapt to the changing regulatory

regimes and then thinking about your architecture at the beginning through those lenses is very important. As an example, when you're operating in 35 countries and highly regulated industries, what needs to be true in a country like India is different than South Africa, which is different than the UK, than the United States. And all it takes is one mistake that's material to mess up a quarter, right? And to have a customer impact. So those things all need to be true. I'd say another lens to this, which is a different

lens

is around the concept of identity. We're now seeing identity or AI agents being rolled out for consumers to make purchases on their behalf. All So it's when I think of what MasterCard is doing and what Visa and PayPal are doing, that is something that's moving very quickly. Now, if you think about the implications, AI agents are becoming a whole new threat factor for companies. And there's a lot of technology and time spent on is this

individual person who they say they are, is their intent what they say it is, and the same is true about businesses. But now that's true about AI agents that have autonomy and the ability to reason and make decisions, you need to actually triangulate those things. Is this AI agent actually representing this person or this business? Do they have the authority to do what they're trying to do? Is this AI agent actually the one released by

that company or is it a spoof? And I'd say the third vector that we're seeing is all the byproduct of this very powerful technology being put in the hands of people with bad intentions. So the fraudsters. So the deep fakes. I fear for the day where I might get a phone call that sounds just like my daughter in a panic situation, right? And we've all started to probably hear those potential stories. We need to be prepared for that, not just in our personal lives, but within our companies as a whole new

threat vector. So data privacy compliance, I look at through at least those three lenses of developments of understanding the product and the identity itself, right? And then what happens with fraudsters, but those are all very important vectors in this conversation.

Jason (15:11)
I'm as concerned about my kids creating agents to start to buy things on my behalf. You know, don't want the Naismith kids doing that either, but you're exactly right. I've actually heard people who are trying to build agents for e-commerce personally, and the purchases are getting denied. Like they can't get through that last, you know, privacy check, if you will, right? To say, are you really who you are or not? Sean, we're on the last question.

Sean Naismith (15:16)
Thank

That's right.

Yeah.

Jason (15:38)
believe it or not, so we're rocking. And this one's kind of a little bit of prognostication, but looking ahead to end of this year and beyond, which things are moving so fast, it's kind of hard to do it, but what emerging trends do you see, whether it's in AI or data, identity that we just talked about, that you think will be most disruptive for, whether it's mid-sized businesses or the private equity firms that we work with that work with these.

know, mid-sized businesses, what do you think they should be thinking about?

Sean Naismith (16:09)
Yeah, so if I was to pick my top three.

I would start with AI moving directly into operational workflows. if the language model is the brain, right, the agents, the body, and those bodies are becoming very powerful at this point, right? And that is now, it's beyond chatbots, right? It's taking that ability to have autonomy and make decisions to actually take actions within organizations. So I'd say in 2025, we started to see that rolled out, not just at the hyperscalers, but at some of the larger financial

institutions as an example, but I think throughout this year and going into 2027, those operational workflows, we will see colleagues that are AI agents, right, that have their own logins and email addresses, and that is already happening. So I see a rapid acceleration into that. That leads to, I'd say, the second key trend, which is going to be around identity and infrastructure becoming critical.

Right, now that you've paired this intelligence, which if you listen to Anthropic, it might be conscious, right, what are we dealing with here? That is now making decisions within our companies in autonomous ways. Again, coming to an understanding of their identity and extending the concept of identity into those areas. Right, we're seeing companies like Okta and Google, right, extending what they think about as far as identity into the agentic realm as well.

third is actually comes back to the foundation, let's say it's going to be data, but govern data collaboration, right? Companies as they start to adopt and apply AI will really realize the value of having proper data ownership and data stewards and clean data pipelines. So big focus on getting your data house and older in order. It's something many companies know they need it to do, but it just doesn't hit that ROI threshold internally and it's never prioritized.

with the ability for AI to unlock the business case, it's probably going to unlock that focus on the data layer. And once you do that, you start to realize the ability to collaborate in the data layer with other companies to get even more value out of the AI that you're running, whether it's an AI-enabled product that you have or whether it's internal for your company. But how do you do that in a privacy-protected way?

I think of privacy enhancing technologies and clean rooms, that's going to extend beyond many of the marketing use cases where it started into many other vectors, right? And having cloud native technologies with built-in capabilities for depersonalization, anonymization, but still allowing linking and matching, I think will be a third key trend that's going to unlock even more value from the AI that we're seeing rolled out overall.

I mean, in summary, when you think of private equity firms, it's, think when you apply those capabilities, even across a portfolio, right, it's going to, those impacts are going to compound very quickly. So I would see those being key drivers here.

Jason (19:07)
Absolutely.

And I think, I mean, the theme keeps coming out. And I think you're exactly right. It's kind of hygiene and governance, right? Which, right? So there's this piece of foundational elements to make sure that, you know, that's the garbage in garbage out, but it's even accelerated now, right? To where we are with the tools. The governance piece to me, and you mentioned, like when initially you think about governance, it's like, okay, you know, is the...

Sean Naismith (19:17)
I do think governance, yes.

Yes.

Jason (19:35)
the federal government going to get involved in this. And I don't think that's going to happen anytime soon. It's really incumbent on the corporations, the companies, the businesses themselves to govern, which I think given the challenge, think is that given the cost is so low and the power is so high that a lot of individuals want to just go, right? They want to take this on.

Sean Naismith (19:39)
Right.

Yes.

Yes.

Jason (20:04)
And so, you know, I don't know if you've dealt with this, you know, how do you put that kind of government structure into place to try and avoid that? Because it's to your point, you know, things can be spoofed or infiltrated or or just information could be getting out that people shouldn't be putting out in public.

Sean Naismith (20:23)
No, you're absolutely right. So I think about this as people process technology, like many of us do. I always like to start with people. And when you think about a committee, as an example, if you're a larger company, many times you need to have a risk committee that either you have an existing one that you're going to increase the scope of to include AI risks or you have a separate committee that needs to be cross-functional. And not all AI use cases carry the same risk. understanding risk tiering, understanding

the

use case, what data is going to be required, who's going to be using it for what purposes, and understanding the risk ranking of that. As many times you could have a high value use case, that's low risk. That's the sweet spot for many companies to get started with. And that's a very powerful team to pull together with the right mandate and the right ability to make decisions. Now you mentioned just wanting to dive in and run fast. I am all about diving in and running fast. My best friends now are Claude and

codecs. And at this point, it's like, you need to have that bounded and in with the right expectations, right? You want teams and team members to learn fast and move fast and understand what's possible, but you need to do that in a safe way, right? You're not just going to say, Hey, here's all of our customer PII information, go to town. Now you're probably going to say, here's a secure environment. Here's the type of data assets that we're going to allow in that environment. And here's the tool set.

Jason (21:21)
Hahaha

Sean Naismith (21:50)
And you can go crazy in there, right? So to speak, you know, can move fast and iterate quickly. And that's kind of what I was getting at with understanding a prototype and a demo, right, versus a hardened production system that really takes into account the workflow, the changes in the operating model that's going to be needed, all of the governance infrastructure that's required and more, right? Those are those are two different ways of thinking. I've seen success running both

of those rails, right? The ladder is going to take longer, right? But the former, you're going to get folks excited, understand the art of the possible ability to innovate and move quickly there. And I would say run both those rails is very effective because then you're not a blocker, right? What you don't want to do is send out a memo to everybody saying, hey, guys, cancel that our company. never doing it.

Jason (22:32)
Yeah. Right.

I don't think toothpaste is not going

back in the tube at this point, right? It's almost like how do you innovate in a structured way, right? So how do we give people that ability to create and build, as you said, but also not to break things, right? Or share things that could be damaging. Well, we went through our questions and as a nod to your great grandfather, was a slam dunk, right?

Sean Naismith (22:43)
The genie's not going in the bottle. exactly. ⁓

Right. ⁓

Yes. Yeah, that's

right.

Jason (23:06)
I mean, maybe if

you want to share to that, like the Naismith name, if people know basketball, this is a legendary name.

Sean Naismith (23:14)
Yeah, it is. Well, it's been obviously part of my life story. I'm a great grandson of Dr. James Naismith, an innovator.

And it's pretty amazing to create a game in 1891 that went viral for its time. I don't know how you go viral in 1891, but it happened. it's quite a legacy. And a lot of lessons learned from how he developed the game. if you've never studied Dr. Naismith and you're listening to this and you like innovation and rapid prototyping, it's a wonderful story. OK.

Jason (23:29)
you

Terrific. So before I let you go, just this is my last bonus question.

It's really, you you're one of our newer advisors on the platform. I'm so thrilled to have you. But as businesses and firms look to engage with you, what do you feel like you are best positioned to help them solve an answer?

Sean Naismith (24:04)
Absolutely. So I would say most of my career.

has been helping organizations turn data and AI into operating platforms and driving revenue and the bottom line. So this usually means three key things, right? The first is unifying fragmented data and identity, right? Breath and depth of experience and understanding that. The second is embedding AI into workflows. And it goes beyond AI, it's innovation in general, right? Is a key part. How to think about innovation within a company

right, and to really get the organic growth drivers going. So much of that these days is filtered through the AI lens because of the promise of AI, but it's innovation in general. And I'd say the third is bringing the experience and expertise to help govern and scale it out, right? So again, people process technology, all are very important. And the goal is to go from AI experimentation and innovation experimentation through full

operating systems that can scale and really drive both top line and bottom line for companies.

Jason (25:10)
Well,

whoever engages with you, and I know there's going to be a lot of companies that do that. They couldn't be more fortunate to have someone like you providing the type of wisdom and experience that you have. So thank you. Well, my pleasure. Great to talk with you. Thanks for making the time, Sean, and sharing all your great insights. Be well.

Sean Naismith (25:21)
appreciate you and the wiserie for the opportunity on it.

Thank you, Jason. You too.