#297 Jeff Lunsford: How Agentic AI Will Redefine Every Digital Interaction

Why will agentic AI redefine every digital interaction, and what foundation do enterprises need to make it safe, trusted, and real time? In this episode of Eye on AI, host Craig Smith sits down with Jeff Lunsford to unpack how a neutral customer data platform like Tealium becomes the control plane for agentic systems. We cover how to collect and unify first party data responsibly, enforce consent and identity across channels, and feed the right context to models so agents can act with confidence in the moment. You will hear how real time profiles, event streams, and deterministic identity power personalization, automation, and transactions across web, mobile, ads, email, and customer support. Learn how leading enterprises are preparing for agentic commerce that could double digital interactions, why governance and privacy must be embedded into delivery teams, and which standards enable safe transactions and payments with agents. You will also hear how to build an "agentic front door" for your business, design guardrails and spending allowances, choose where to run reasoning and inference, and measure impact with metrics like conversion rate, ROAS, CSAT, and cost per resolution. ​​Stay Updated: Craig Smith on X: https://x.com/craigss Eye on A. I. on X: https://x.com/EyeOn_AI

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  This is happening so fast, so the ways this is going to evolve, we can't really anticipate beyond a year or two. And I think it's good that the velocity of spend and capital formation is all going to get less friction in capital formation, easier to start companies, easier to engage visitors, easier to monetize yourself as an individual. It's going to be wild. It's going to be fun. Think of us as running inside your four walls, but yes, it runs in the cloud. And so you're just going to use our software to collect your first-party data, define your attributes that you want to, you know, your own language to enrich this data with. Helium's original mission was to help the largest companies in the world

• 0:45Speaker 00

  collect and enrich and then create experiences around their first-party data, their customer data. So as software eats ate the world and the customer experience was created by more and more specialized software packages, each of those software packages had its own data set around the customer. And we felt the industry needed a neutral provider to come in, kind of the way Tibco did in the mainframe era or WebLogic did in the app server era, and be a neutral provider that connected to all these upstacked systems and to all the downstacked transactional systems, and listened to the event-level data from customer experience systems, resolve that event-level data around visitor profiles.

• 1:30Speaker 00

  So we basically turn machine data into human data and then allow companies to enrich those profiles with their own language, right? So a big pharma company might have different attributes they would assign than maybe a big airline would or a big bank, you know? And you know, Helium works with 10 of the 20 largest banks in the world, 6 of the 10 largest telcos, 3 of the 5 largest airlines. So these companies have customer data flowing everywhere. And, you know, so we'll perform that identity function, we'll allow them to enrich that data. This all happens, Craig, on a high-performance, real-time platform. And it all is on a governed and trusted platform because, you know, we've had privacy regulations around the world.

• 2:14Speaker 00

  And so now a big airline, big pharma, you know, big hotel chain, uh, big e-commerce company can collect all this data, follow privacy rags and then use trusted and compliant customer data to fuel customer experiences. And now this was all happening in the martech, ad tech, CRM landscape prior to AI. It turns out that as these big companies embrace AI and start using AI to fuel customer experiences, this is the data set that they need to populate the context window, right? To give you this real time context, the behavioral data around

• 2:54Speaker 00

  your customers' activity that the LLM needs to understand, right? And the LLM likes structured data, it likes label data, and that's exactly the, the stuff that Tealium produces or helps our customers produce. So that's kind of a long introduction, but um, it's all about basically a trusted platform. And whether companies use Tealium or not, they have to solve this mandate, right? They have to follow privacy regulations. They really need to get this customer visibility, uh, visibility into real time customer data because that's that real time data gives you the highest intent signal, whether you're an e-commerce platform

• 3:36Speaker 00

  or travel platform or whatever. And that those intent signals are, are what that data is, what the, the models can use. And, and by the way, this isn't always an LLM. It could be a fraud algorithm model. It could be a um, cat boost model. It could be an LLM. Um, but you know, basically more data equals better outcomes with AI. But it's not just a completely unlimited fire hose of data. You know, we now, we're now learning in AI that giving too much context can, can lead the models to wander, right? So, uh, you know, one of our founders talks about, um, instead of flying the needle in the haystack, how about, uh, creating smaller haystacks?

• 4:21Speaker 00

  And so, you know, you kind of want to give that tighter, higher resolution profile to the model so that it can do its thing.

• 4:30Craig S. Smith?

  Yep. And let me ask some questions about that. That's, uh, that's interesting. And I haven't sort of broken down this data aggregation and.

• 4:40Craig S. Smith?

  Rationalization process. So how many different channels can you pull data from? And how many different formats, not specifically, but?

• 4:52Speaker 00

  Sure. So basically, anything that's digital, so the web, mobile apps, call centers, point of sale, set-top boxes, over-the-top television, kiosks, ATMs, store data, you know, from Wi-Fi routers and stores, basically any kind of event-level data.

• 5:17Speaker 00

  And we take that, again, that machine-level data, and we resolve it then around identity so that you get an actual visitor profile. And it all happens in real time. So it's a big, you know, compute load at real time, and then we allow that profile to be available back to, you know, like, let's say, put AI aside for a minute. If you're just doing classic online advertising, so Meta, Google, Snapchat, Pinterest, Reddit, they all have conversion APIs. And when your customer's visiting, you can send them data in real time, and you give them feedback on what is converting.

• 5:59Speaker 00

  And then they actually use AI behind the scenes to basically help companies dial in the focus for their audiences that they're targeting. And we're seeing an average of 26% improvement on return on ad spend for companies using these conversion APIs. And so, I mean, you know, you talk about there's 500 billion or whatever of digital ad spend, just in US and Europe this year, probably. So 26% of that is a pretty big number. And then when you add AI to it, you know, we're seeing even tighter and more impressive outcomes. Like, we had, you know, an auto dealership that was using Snowflake and Tealium

• 6:43Speaker 00

  to get their data act in order, and, you know, they went from a 16% conversion rate at the bottom of the funnel, people that bought a car, to a 30% conversion rate, right? So more than doubling.

• 6:57Craig S. Smith?

  Yeah, and and when you say that all that data of coming in, it's it's got to be put into a machine readable format. Right? If you're getting call center data, how how does the system convert that in? Is it converting it into vectors or or or is it using natural language? So there's separate natural language.

• 7:26Speaker 00

  We we have, think of us as before, like the vector store or the graph database. Now, for AI purposes, we definitely are are helping vectorize this data and using graph databases to show the relationships of the attributes. But prior to that, just the core tooling platform is a deterministic, not a probabilistic. Um, model and we just understand all, you know, we have 1300 connectors that we call them connectors. But where we can ingest inbound data and part of building one of those connectors is understanding the data format you're going to get from that source. Right? From a, you know, Samsung TV or from a mobile app or whatever.

• 8:09Speaker 00

  And then you do, you map field by field, these elements that are going to come in on the event level data. And then our customers can use, we call it an identity key ring, but you know, multiple different.

• 8:23Speaker 00

  Dimensions can be combined in identity key ring to resolve around identity and to turn this event level data and associate it with a human. Right? And so it could be email address, it could be device ID, you know, it could be some other signature. URLs, all that kind of stuff. And so that is all hap- that all happens deterministic and this is a pretty key point, Craig, because there are a lot of folks out there talking about probabilistic matching and things like that. But but when you, when you look at global privacy regulations.

• 8:57Speaker 00

  You cannot have probabilistic anything going on because if a customer, one of the most important things we do is we capture the consent status of a customer, of a visitor to any of these digital, you know, channels. And then we propagate that downstream to all the systems that can benefit from knowing that consent status, right? Because this company over here can tar- if they target someone who just opted out, they can get fined. And, you know, it's very important that you don't start saying, oh, we think this person might be, or these events might come from that person. So we're probabilistically going to stitch them together. And then you

• 9:42Speaker 00

  start targeting that. And if your probabilities were wrong, you are now in violation of privacy regs, right?

• 9:50Craig S. Smith?

  Yeah. And all this is happening in real time, in the cloud. I mean, are you a SaaS platform?

• 9:58Speaker 00

  Yes, we are. So the way you think of it is, Telium, I would say like, we're just software. Like, think, you know, when you're talking to a customer, think of us as running inside your four walls. But yes, it runs in the cloud. And so you're just going to use our software to collect your first party data to find your attributes that you want to, you know, your own language to enrich this data with. And we think of it kind of in three tiers, right? So you've got all your customer experience systems that are typically outside the firewall. Also now, they're mostly cloud based systems. Every once in a while, you still have a big bank or something running a CMS on-prem or something like that. Then

• 10:37Speaker 00

  you have Telium, which creates this trusted data layer. And think of it as kind of like a high performance cache that sits at the edge of the network. And then in the back office, you will have your point of, I mean, sorry, your transactional systems, like a DDA system or credit card processing system. And you might have like your data warehouse, right? Like a CDW, like a, you know, Snowflake or Databricks, or Redshift from Amazon, or Google, Big Tables, and then-

• 11:12Speaker 00

  Our big query, sorry. And then those are what I, what we call kind of like the stateful at rest backend customer data systems where you want to run your heavy workloads and your heavy analytics and things like that. We are the high performance cache where when someone is visiting one of your properties, you want to have that profile at the ready. So you can populate, you know, send it to the LM in an AI use case or send it to the media property in an ad tech use case in real time with the most up to the millisecond behavioral data, right? Because it's what I'm looking at in the current session gives you much more intense signal than what I was looking at two days ago or two years ago. Right?

• 11:58Craig S. Smith?

  Yeah. Wow. And presumably you, you've been developing your, your, your platform or your software as AI has gone through these successive

• 12:14Craig S. Smith?

  advances. And right now we're, we're in the agentic phase. I'm, I'm, I'm a little bit of a skeptic on agentic AI. I think there's a lot of, what are they calling it? Agent washing where there's not necessarily some real agency there. But can you talk about how your system has adapted to those changes?

• 12:45Speaker 00

  Sure. So first of all, yes, there's probably a lot of agent washing going on, Craig, but, but you know, one of my favorite science fiction writers has a saying, which is the future is already here. It's just not evenly distributed.

• 13:03Craig S. Smith?

  Right. Yeah.

• 13:04Speaker 00

  And If you look at, for example, in crypto, there are already entire agent economies like on virtuals and pump dot fun where agents are interactive with other agents and you know, pump pump has 750 million a day of activity and virtuals has 125 million per day. It like their their own economies. And so what we see, we believe, you know, and in this year, by the way, you're already going to have 136 billion of agentic commerce in 2025. And that is predicted

• 13:45Speaker 00

  to go to 1.7 trillion of agentic commerce in 2030. So

• 13:51Craig S. Smith?

  Can I, can I just stop and have you, I'm not familiar with pump or with the the other platform?

• 13:58Speaker 00

  Virtuals.

• 13:58Craig S. Smith?

  And when you, and when you say agentic commerce, just give me an example of a transaction that you're talking about.

• 14:06Speaker 00

  Yeah. So a transaction is like pump dot fun. They're trying to compete with Twitch, where a creator can go on to pump and do their stream, just like you would with your your podcast. And they get paid based on activity, how many people are watching them in the pump token. And, you know, it's, it's kind of wild, but it is going. That's, that's the the bleeding edge. Now, the mainstream is going to grow rapidly because we see the biggest players in commerce and media promulgating protocol standards that are either being developed or are in

• 14:48Speaker 00

  production. The most high vis one so far was the OpenAI agent commerce protocol that they just they and Stripe announced this week. And I went on to OpenAI and I, you know, shopped on Etsy and went all the way to the shopping experience still within the OpenAI. So you've got this agent commerce protocol, ACP. We all know what MCP was, model contact. Also, I say MCP basically was giving agents hands, you know, so, so giving them the ability to go, do something like maybe make a reservation, but it didn't give them the ability to. Do a financial transaction. So now you have all these different protocols coming at you.

• 15:33Speaker 00

  So the Ethereum group has promulgated ERC 8004 as a standard for agent identity. So this is a new trust layer that will go on. And then you've got Coinbase, which promulgated X 402, which is a payment standard. And then you have Google, which, which promulgated A2A, age of protocol. And they then have agent commerce protocol, which is inclusive of and supportive of the X 402 from Coinbase. So Google and Coinbase basically setting a standard. You've got OpenAI and Stripe setting payment standards.

• 16:16Speaker 00

  And so agent to commerce is simply where I can have my agent, like ChatGPT or any other,

• 16:25Speaker 00

  go do a transaction for me in, in, in, in the future. So there's a couple of things there. So I have spent 30 years in fintech and I'm on the board of the bank and helped, helped the team launch the world's first internet bank back in the late 90s. So, you know, the, the intersection of, of technology and finance, which is now known as fintech, has been going on a long time and it's very fascinating to watch. But, you know, what we saw over the last 30 years, Craig, between the internet revolution and the mobile revolution, we're seeing even more change happen now over the next four years than we saw over the past 30. And because all the rails are built, you know, and so we believe there will be 1.7 trillion and other,

• 17:09Speaker 00

  you know, third parties, not Tealium, you know, are predicting that by 2030. That's about, it's a little over 20% of global digital commerce. So, I mean, if you're that, I think that's kind of a mid to bearish forecast. If you're bullish on it, it could be half because, you know, like Gartner saying,

• 17:30Speaker 00

  Ninety-five percent of B2B buying journeys in next year, 2026, are going to be initiated through agents. Now, that sounds kind of crazy, but if you think about it, think about your own behavior, right? How, how much are you now starting it? This is B2B, think about it. So I am, you know, I'm here at Telium and I want a new HR system. I'm not going to Google it anymore. I'm going to ask Chad GPT what the leading vendors are and what, which is the best fit for Telium. And so I could see where probably that 95% next year is pretty accurate. So at the top of the funnel, the shopping or the searching will, will start with agents.

• 18:16Speaker 00

  And so it's just logical connection that these intelligent agents, once they have like an identity trust layer, like what ERC 8004 provides and payments capability, what like A2P or A2A provide,

• 18:33Speaker 00

  then we are going to see agentic commerce. And so we're preaching to our customers. We have about 700 customers around the world that, you know, this is coming. And so you need to build agentic front doors to your business, right? And our belief, again, having watched the web and the mobile for the last 30 years and how that changed everything in anticipating what's going to happen over the next four years. Our belief is that your current, let's say you're a company that has a million digital interactions a day to day. Okay. So that's your, that's your total digital customer engagement number.

• 19:17Speaker 00

  Well, in 2030, we think you'll have 2 million, double, double the number of digital interactions. We think the current million aren't going to really go away because, you know, human behavior is, is, you know, it's a, it's a slow slope. It's not an event. And so you're, so you need to preserve everything you're doing to be successful and competitive with the current digital world. The other million are going to be agentic, right? And so you need to have that agentic front door. That, by the way, it doesn't look like a website, right? It's not going to be agents coming to use browser interfaces.

• 20:02Speaker 00

  Yes, agents can use browser interfaces. So yes, that your website maybe will work to convert some visitors. But if you have a high-performance agentic front door that supports all these protocols, the trust, the identity, the payment protocols for agents, then you're going to get more of the business than your competitor who doesn't have that. Right? So, you know, we're preaching, you're running the risk. If you're not thinking about this now and architecting for it right now, you run the risk of losing a generation. Right? So, you know, 10, 15 years ago, if you did not launch a mobile app, you'd lost a generation. You don't have any of my kids that are in their twenties and early thirties.

• 20:47Speaker 00

  And in 20 years ago, if you didn't launch a website, you lost a generation. You lost, you know, my generation. And so, and even before that, if you didn't launch a voice response system, you lost a generation. So companies will lose an entire generation if they do not lean into this. Now, what's, what's very interesting, Craig? So when you, when you support enterprise companies that have to follow global privacy regulations, while we are evangelizing, they lean into it. They also now increasingly are setting up basically AI risk committees or AI approval boards where before you do anything.

• 21:28Speaker 00

  With AI, you have to kind of send it up to the board. Um, and that's actually slowing big companies down and giving smaller, more agile companies or more risk oriented companies an advantage. And so now what we're seeing is kind of, there was this concept, you know, of privacy by design. Um, that kind of came 10 years ago and when you, you know, Europe started with GDPR. And we think the same, the best practice now i- in the big companies is, is they're actually taking these people from the AI. Uh, approval board and embedding them in the development teams that are building these projects.

• 22:10Speaker 00

  Because, you know, it's kind of like having an embedding, an embedded media, uh, person like you out on the, out on the battle battlefield, right? So you just, that helps bigger companies speed this up. But I, we believe it's just what I call a strategic mandate, you know, embrace it and build for it, or you'll miss a generation.

• 22:30Craig S. Smith?

  And, and you're, so you're talking about not just preparing training data, you're talking about governance and, and, and how a gentic needs real time policy enforcement, is that right?

• 22:47Speaker 00

  Yeah, it's governance and it's, um, so the training data you can't really use unless you've anonymized it. You can't really use customer data in your training because of the privacy problem, right? Cause those customers may opt out at any time and then that paints your whole model. So for customer specific AI use cases, you need to wait and populate the context window at time of inference. And, um, that is mo- the, the data that I'm talking about where this high performance trusted cash can, that can give you the, the visitor profile at time of inference comes in really handy.

• 23:34Craig S. Smith?

  Yeah, that's, uh, it's amazing how this stuff is developing and, and, um,

• 23:43Craig S. Smith?

  Uh, come of of consumer data platforms, uh, uh, uh, are, are they, uh, are you providing, uh, dynamic control systems, uh, for this AI first world that we're looking at?

• 24:03Speaker 00

  So, um, yeah, the, the product category we invented back in 2013 is called customer data platform. And it

• 24:11Craig S. Smith?

  Oh, I'm sorry, not consumer. Yeah.

• 24:13Speaker 00

  Yeah, that's why it, it, um, it was such a good idea that, you know, four years later, we had 130 competitors, other companies saying, oh, we do customer data platforms. So, um, you know, the, I, the simple idea was what I started with. Just, okay, you got all this data everywhere about your customer and you need to resolve it into visitor profiles and be able to, uh, leverage it, but also govern it. And so then when AI kicks in,

• 24:44Speaker 00

  and if you think about this agentic world, you know, prior to AI, there was this concept in Martech called journey orchestration. Where you could send a customer down different paths and, and what we always preached was, look, you can't do some beautiful whizzy wig customer journey map, you know, in a software interface because you just run out of it. There's too many permutations of possibilities on most digital. And so what you really do is you orchestrate customer journeys by orchestrating data. And so when a customer does this, you enrich the profile, then you have rules upon which the customer, you know, you can

• 25:29Speaker 00

  enhance the customer experience based on what they just did. And then you have more rules and more rules. And so through data orchestration, you can do amazing journey orchestration. And so thinking about agent orchestration is sort of the next logical step, right? So it's now, whereas these business rules that you were doing, um,

• 25:55Speaker 00

  Data orchestration with and journey orchestration with were deterministic. Within the agent world, everything is a little probabilistic. Now, it's getting more and more tight, right? Like, um, we, I tell my team all the time, like, you have to fly the airplane looking out of the the front of the cockpit. I was a naval aviator way back and, you know, it's like you don't fly the airplane looking in the rearview mirror. And, and so you have to assume this technology is getting better and better. So three years ago, we saw 15%

• 26:33Speaker 00

  hallucination rates. And so all the naysayers were saying, well, you can't use AI 50. Well, it was, it's now less than 1% in 2025. You know, it'll be less than 0.1% 2026. So you have to assume these things are getting closer to deterministic.

• 26:52Speaker 00

  And what will happen is you'll have an agent that, you know, you'll, you'll ask it a question. Here's an example. So one of our customers did this, they, they said, okay, so this visitor is on our website. This is a large global chip company. They're on our website and, um, they're, they've looked at these three products and our 80,000 parts SKU. Speculate for me, um, what industry they work for, what kind of project they're working in. And in what three other SKUs our sales team should call them about tomorrow to include in their project research, so kind of a B to B use case. And then they would ask the LLM that the LLM would answer with the three parts and they'd put that in the CRM.

• 27:36Speaker 00

  And then the sales rep would get a notice, right? So of, okay, now call this person. Now it's very simple to say, just, let's just put that back in the website in real time right now too. Right? You know, and so that kind of thing is what we're seeing, you know, very simple, but very high value AI use cases with this data.

• 28:00Craig S. Smith?

  Yeah.

• 28:00Speaker 00

  That could, that could then another, that could be the first agentic output. And then there could be another one and another, another one, if you think about it. And that would guide the customer journey.

• 28:11Craig S. Smith?

  But, but who's can, is this all?

• 28:15Craig S. Smith?

  Being orchestrated by a reasoning agent or or do you have sort of rules in place? What's the control mechanism for for for deciding what you're going to send to the website? What you're going to send to, you know, the CRM and that sort of thing?

• 28:37Speaker 00

  Yeah, so the control mechanism pre AI is we have rules and, you know, we'll say, you know, if a customer abandons their shopping cart with three items in it, then you would add them to a retargeting audience on Google or Criteo or whatever, right? And then, or if a customer converts and buys something, then you would suppress future media. I mean, how many times you've been retargeted for stuff you already bought? You're like, these people are wasting their money, especially for a long considered purchase. Like, I just bought a leased a car. Well, that's going to be a three year lease. So why, you know, why waste money advertising it could be for the next two years at least? And so those things were rules that our customers set up.

• 29:20Speaker 00

  Now, and again, the action then happens through these connectors we talked about. So now it'll simply be, you'll ask a model, you know, what you should do. And you can ask the model any question you want, as you know. So you could say, hey, given, you know, this audience profile that this person fits in and their recent activity, you know, what is the next best product recommendation we could put to them the next time they visit or even on the next page in session? Right? That's where when you get into the real time, it's really the in session change of journey.

• 30:02Speaker 00

  And if you take this to its logical conclusion, Craig, every like the the concept of an audience, we think is dead. We've been preaching that for a long time. Basically, every every visitor is an audience of one and you're going to create a personalized experience for them. And then ultimately, and I don't know if this is one, two or three years, if you think about like what Sora can do and all these other real time content generations like. Every visitor should have their own visual and audio experience, right? Now, remember, that's still the first million interactions. Now, for the other million interactions where their agent comes on their behalf to transact, that'll be different.

• 30:46Speaker 00

  That'll be through, you know, APIs and there'll be data flowing back and forth, but it will still, you can still determine and modify what you send to their agent that is working on their behalf to drive conversion. And, you know, a conversion isn't just an e-commerce term, right? We work with large banks, large lenders, uh, uh, just filling out a loan application is a conversion, you know, that's super high value for a bank.

• 31:19Speaker 00

  And any, almost every business, um, in, in big pharma, where, you know, you're, you're dialing in different content to different audiences, right? So if someone, and by, by the way, as controlled by geography. So if you're going to tell a cancer patient, you're going to give them different con- content, then you are a cancer researcher, then you are a cancer doctor, then you are a pharmacist, right? You're going to have the different content to each of those audiences, depending on who they are and what they're looking at.

• 31:58Speaker 00

  And so super high value use cases like that, um, that then the pharma company can actually kind of connect the dots on, okay, when I put this in and I started dialing in my content, we saw a script lift of, you know, 32% or whatever in Spain, when we started targeting content for this new oncology drug we're launching in Spain.

• 32:21Craig S. Smith?

  Right, but, but that moment when the system is figuring out next steps, is that a human operator that's looking at the data and entering a prompt? Or is that a reasoning?

• 32:36Speaker 00

  That's a reasoning engine and, and, and that's the point of it's just going to get faster and faster.

• 32:41Craig S. Smith?

  Right.

• 32:42Speaker 00

  And, you know, obviously inference, um-

• 32:47Speaker 00

  Everybody is working very hard to make inference faster and inference is getting, you know, 30 times faster. The latest Nvidia chips, the inference is 30 times faster. And that's not just optimizing at the chip, it's optimizing through networking and algorithms. And that's 30 times faster in one year. Now, think of that compared to Moore's Law. You know, Moore's Law is, what, doubles every two years or something like that? And, and so the, the exponential pace at which AI

• 33:28Speaker 00

  performance is improving is kind of hard for the human mind to comprehend. And so inference, that's what I mean by it. It's pretty clear that we're going to go to where, where reasoning right now could take 30 seconds. Reasoning will take point three seconds and you'll still get all the benefit of that reasoning.

• 33:48Craig S. Smith?

  Yep.

• 33:49Speaker 00

  And there are companies, hardware specific companies like Grok with the Q and Cerebras, which is building a super large chip that are basically optimizing just for inference. And then there's Nvidia and others that are, you know, optimizing for training plus inference. But it's all just going to get faster and faster and faster. Where the end state is you get your personalized reasoned output as a consumer that is only for you.

• 34:21Craig S. Smith?

  Yeah. I mean, we started talking about governance and, and governance. And, and you were saying how it's slowing some companies down and that's creating opportunities for others. But I can, I can see with this system as it gets faster, governance becomes less and less of a bottleneck. But can it accelerate innovation? It seems to me it would still hinder innovation.

• 34:53Speaker 00

  Yes. I mean, governance and privacy compliance.

• 34:58Speaker 00

  It definitely slows you down. I mean, if you just look at drug approval in the USA, like, you know, the three phases and how long it takes. And even after a phase three trial, it might be a year before the FDA does their review and approves a medication. It's like, why does that take so long? So, you know, of course, because of safety and but everything should happen faster, right? Drug approval should have faster customer experiences and the enterprise should happen faster. And the way to speed that up is to make it programmatic, right? And have an air traffic control layer for your for your data or air traffic control for your data layer. And if you're going to impose governance, data governance, it's

• 35:41Speaker 00

  way easier to impose it at one centralized choke point, if you will, than it is in a bunch of different distributed places. And that is one of our value propositions. Again, whether you do this with Tealium or or build it yourself, having all of this, the data that is subject to privacy regulations, right, which is your customer data. So you don't you don't need your supply chain data and other data, but all of your customer data needs to come through here. That's where we we track consent, we understand the provenance of the data, and then we can activate the data by activate a customer insurance by sending the data out to all these different systems. But that's what they consent known

• 36:25Speaker 00

  status. That all can happen in real time. It doesn't need to be a batch process. And so you dramatically decrease your regulatory burden when you basically adopt this architectural approach of having this neutral trusted data layer.

• 36:47Craig S. Smith?

  How deep is the penetration of this technology? It's funny, you mentioned, you know, buying something and then for the next week you get ads served to you. I mean, I, you know, it seems to me that. That should have been solved a long time ago. As you're saying, it is solved, but it's it hasn't penetrated. How, from your point of view, uh, how, I mean, I guess for Telium is kind of an unlimited market.

• 37:18Craig S. Smith?

  And it's going to take a long time to saturate that market. Where do you think the the economy is on that?

• 37:26Speaker 00

  So, you know, I think, again, even pre-AI, we would, we still run into companies where they're still using week old audiences to run email campaigns. And these are the biggest companies in the world. And then, you know, when you come in and you change that to real time, you know, you get, you know, fifty plus percent improvements in in email open rates, right? Um, we just had a published case study with Legal and General over in the UK, largest asset manager there. And they saw fifty-four percent improvement and, you know, open rates just by doing stuff with with Snowflake and Telium.

• 38:05Speaker 00

  And so we we still see a lot of pre-AI low hanging fruit. And um, the more regulated the industry, probably, you know, the slower they've been moving. And but that's why you you get these new, you know, um, uh, new new banks where they don't have all of the old school legacy stuff, or, you know, the Affirms with Max Levkin, who's, you know, tech genius, right? Um, and, you know, he's just moving faster, you know, um, or Stripe, or Square. You know, they can just move faster than the legacy institutions.

• 38:49Speaker 00

  And the legacy institutions are, however, you know, have these amazing businesses, these amazing franchises built on trust. And so the whole idea is how can we help them move as fast as the fintechs? While still maintaining, you know, keeping their security and compliance people happy. And it is doable. The whole key is getting high enough in the organization that you get executive sponsorship for one of these projects. Because I'll tell you that the biggest impediment to adoption is not technology. There are no technical barriers.

• 39:31Speaker 00

  It is organizational confusion or turf wars or friction. And well, who owns this customer profile? Is it the marketing team or is it the customer service team or the sales team? Or, you know, and then it's all of this, who owns what that slows things down. But if you get to where the C suite of a company is pounding on the table saying, you must do this, you know, let's, let's, uh, break through the bureaucracy. And one large bank we're working with has, you know, the CEO gets, uh, gets briefed every week on what he calls his bureaucracy busting projects where I wonder every week.

• 40:12Speaker 00

  If we're making progress here, because you know, we've just over the years built up too many layers of and confusion of ownership. And so that's really what it takes. It takes, you know, the, the, the will to embrace the risk smartly. And evolve your institution and your data practices at smartly. So that you can embrace this because again, you run the, if you do not have an agentic front door. You are not going to get my daughter, 32 year old daughter. She has three kids. She's talking to Google and chat GPT on her phone while she's, you know, got the three month old and her other arm. And, and if you don't get her while she's thinking about something, you're never going to get that conversion.

• 40:56Speaker 00

  Like the next hour, she's on to something else and another kid has a crisis or whatever. The next day, forget it. You're never going to get her. So you literally will lose a generation if you don't solve this, this, um, it's both a challenge and an opportunity for big companies of building what, what I call agentic front doors.

• 41:14Craig S. Smith?

  Yeah. And, and where do you see, you, you obviously see agentic AI.

• 41:24Craig S. Smith?

  Continuing to expand into the economy. It's, where, where do you think it's headed?

• 41:31Speaker 00

  Just, you know, like I said, there's, you know, let's say 136 billion this year, 1.7 trillion in 2030. You'll see it, you know, and, you know, the future's already here. It's just not evenly distributed, right? So, so you'll, you see it in crypto first and then, you know, you'll see it in low valued, low risk transactions. Right? And then you'll see it on bigger and bigger transactions. I think it's going to be a while before my agent could go book a home loan for me. Right? But I do think my agent will be able to, you know, book a, an air travel flight. So what's happening with these protocols is

• 42:13Speaker 00

  an agent will be, there's, there's what ERC 8004 does is it creates like three levels of trust. And it creates an identity layer for agents where my agent will get an identity token. And I, as that agent's owner, will give it an allowance, a limit. Okay. You can go spend up to a hundred dollars for me. And that agents, you know, so whoever's coming from the other side, and by the way, Visa and Mastercard, Visa has Visa Intelligent Commerce. And Mastercard has Mastercard Agent Pay, right? So, so this is happening.

• 42:53Speaker 00

  And so my agent will have an allowance, basically, to go do, you know, let's say up to a hundred dollars or up to a thousand dollars, whatever the number is. In transactions, no bigger than, you know, 50 bucks at a piece or whatever. And that's how you'll kind of control the risk, both me as a consumer and the, the entities out there. But it's happening. It's going to happen very fast.

• 43:19Craig S. Smith?

  And, and when you talk about, oh, the front doors, you're not talking about APIs. I mean, the, the agent, if it's instructed to go buy a plane ticket.

• 43:33Craig S. Smith?

  Will it go to the URL first and then will there be something in the URL code that will connect the agent to uh its interface or or will it does does the agent have to call an API for that specific business?

• 43:52Speaker 00

  I think you you know smarter people than me are architecting these things, but I I haven't written code in 30 years. Um but I did one at one time where I didn't write code, but I think, you know, big picture, what will happen is it will be an API that probably it will get sent to the same URL, which will then resolve on this. Is this a human or an agent? And if it's an agent, it'll expose the data layer and you know, let's say the inventory available and you know, all this other stuff through an API. So so the agent will hit the front door, go, okay, I'm sending the agent this way through an API. So I don't have to render HTML pages and all

• 44:33Speaker 00

  that stuff. And it'll all happen at lightning fast speed, by the way. And so the whole agentic journey. So once they've entered the the store through that API, you know, they can do many, many different things on my behalf. Like go go research a great trip to uh you know, Playa del Carmen, Mexico. You know, I like boutique hotels. I want to swim in the cenotes. My wife and I want to have dinner, you know, finds us some cool spots. I want to see the Mayan ruins and my agent can go do all that, right? On any Expedia or any other, you know, website, but let's not call it

• 45:15Speaker 00

  a website now. Let's call it just a digital travel service provider and come back to me and go um either depending on how much leash I've given it. Okay, I went ahead and booked it for you or here are the three packages I found for you. Uh which one would you like me to proceed with? And you know, the way like um as these guys are fine tuning their models right now, like OpenAI a lot of times and Grok and others will say, which one of these answers do you like better? You know, same thing.

• 45:42Craig S. Smith?

  This is going to speed up the economy, right? At least the.

• 45:48Craig S. Smith?

  The transactional economy.

• 45:50Speaker 00

  Yeah.

• 45:50Craig S. Smith?

  And uh, what does that mean for for society is that? I mean, I I talked to my kids, you know, I remember when, you know, everything was typewriters and you needed carbon copies if you were gonna CC people. Uh, you know, you just think about how slow things were back then. Uh, and and yeah, things are fast today. It gives us individually a lot more free time. But how do you see that that in five or 10 years?

• 46:25Speaker 00

  Higher velocity money, I think most economists will tell you is a net positive. Um, and you know, I I'm like you, Craig. I remember when I started in bank technology in 1994, writing on acetate on overhead projectors, you know, to do to do strategic planning sessions with banks. It's like, I mean, there so back then, if I in 1994, if you told me I'd be sitting here with you, recording a podcast, you know, I I would have gone, ah, you're crazy, you know? Um, and so we can't even conceive. That's what I meant by it's taken 30 years for the internet and mobile. Well, it's only going to take four years for 20% of digital commerce to be agentic.

• 47:09Speaker 00

  Like, so this is happening so fast. So the the ways this is going to evolve, we can't really anticipate beyond a year or two. And net net, I think it's good that that the velocity of of spend and and capital formation is all going to get less friction and capital formation, easier to start companies, um, easier to engage visitors, um, easier to monetize yourself as a as an individual. Um, so it's gonna be wild. It's gonna be fun. Um, and you know, we just want to help all of our customers do it in a safe and compliant manner. Um, but we want to be, we want them to be able to dazzle their customers with these amazing AI-powered experiences. And um, that's the path ahead.