In this episode of Consulting The Future, I’m joined by Julie Teigland, Global Vice Chair – Alliances & Ecosystems at EY, to unpack the growing role marketplaces are playing in enterprise AI adoption. At a time when organizations are overwhelmed by tools, vendors, and bold claims, Julie offers a pragmatic view of how marketplaces can collapse complexity rather than add to it.
We begin by exploring why so many AI initiatives stall between pilot and production. The issue, as Julie sees it, is rarely a lack of technology. It is the friction between tools, governance, integration, and operating models. Marketplaces, when designed well, bring together pre-vetted solutions, integration patterns, security signals, and governance frameworks. That coordination reduces the invisible work leaders often underestimate, such as vendor risk reviews, compliance checks, and architecture validation.
Julie is candid that not all marketplaces are equal. The ones that create value are designed around business outcomes, not product listings. Fraud reduction, supply chain resilience, or industry-specific cost transformation efforts require curated ecosystems. That means benchmarks, peer usage patterns, reference architectures, and assurance around regulatory alignment. For executives facing board-level scrutiny, that difference matters.
We also dive into risk and accountability. Julie highlights how many organizations focus on model ethics and accuracy, yet overlook integration risk. Once AI is embedded into workflows, the questions shift. Who owns the decision? How are outcomes logged and audited? What happens when multiple vendor systems interact? This is where governance can break down, especially at scale.
Through alliances with leaders such as Microsoft and SAP, EY is helping shape how AI solutions are vetted and deployed across industries. Julie shares how EY applies a “client zero” approach, using its own marketplace internally to test value, resilience, and governance before scaling to clients. With 70,000 clients, 668,000 servers, 13,600 workflows, and 269 million transactions processed daily, the system is tested constantly under real operational pressure.
We also examine the influence of the EU AI Act and the contrast between innovation-first and responsibility-by-design approaches across regions. Julie argues that speed and governance are not opposing forces. True acceleration comes from industrializing governance rather than bypassing it.
If AI success depends less on technological brilliance and more on clarity, orchestration, and trust, then marketplaces may be evolving from simple app stores into operational control planes. The question is not whether ecosystems matter, but how deliberately you design them. So as AI adoption accelerates inside your organization, are you simply adding tools, or are you building a marketplace strategy that can withstand scrutiny and scale?
Useful LInks
Connect With Julie Teigland
[00:00:04] Why do so many AI initiatives stall after promising pilots, even when the technology itself seems sound? Well, across industries, leaders are investing heavily in AI, but progress often slows when tools, data governance, operating models and accountability all fail to connect. But the challenge is very rarely the model. It's more about the complexity that builds up around it.
[00:00:29] So today's guest brings a rare ecosystem level view of how complexity can be collapsed. Her name's Julie. She's the Global Vice Chair for Alliances and Ecosystems at EY, where she leads the global strategy for scaling technology enabled value through collaboration with hyperscalers, AI native platforms, industry specialists and the list goes on and on.
[00:00:54] But in the episode today, I want to talk about how AI marketplaces are evolving from those simple app catalogues into something far more powerful. Because Julie will explain how curated marketplaces can reduce friction, lower integration risk and help organisations move from experimentation into production, but with confidence.
[00:01:19] And we'll also dig into why governance tends to break down at the scene where AI meets real workflows, and how ecosystems can embed responsibility, traceability and trust right from day one. So with regulatory pressure rising and the EU AI Act reshaping expectations around accountability,
[00:01:40] today's conversation will hopefully offer a more grounded look at exactly how leaders can scale AI responsibly without slowing innovation to a crawl. But enough from me. Let me officially introduce you to Julie right now. So a massive warm welcome to the show. Can you tell everyone listening a little about who you are and what you do?
[00:02:02] My name is Julie Tegland. I'm super honoured to be on the show, Neil. I lead Global Alliances and Ecosystems for EY. Well, it's a pleasure to have you join me today because you're someone that sits at the centre of how enterprises work with partners. And at a time where everybody's chasing AI or agentic AI at the moment and agents, etc., I'm curious from everything you're seeing,
[00:02:26] why are marketplaces becoming such a powerful way to reduce friction and move AI adoption forward so much faster? Because I think we've seen a lot of projects get stuck in pilot phase, struggling to go into production, struggle to scale. But in this area, it seems to be moving a little quicker. Tell me more about that. The marketplaces have made a real difference because effectively they collapse complexity. AI adoption, it's not really failing because of a lack of technology.
[00:02:56] It fails because leaders really struggle to connect really the tools, the data governance and the operating models to that. When you look at a marketplace, it can really reduce the friction by bringing together pre-integrated or pre-vetted and increasingly interoperable solutions into one single environment.
[00:03:20] Effectively, it's giving you a little bit of choice, but with some guardrails, having some curated listings to help providing faster integration, procurement efficiency, of course, and maybe some better observability from day one with some built-in monitoring for usage, cost, drift and performance. So marketplaces, they're really changing the AI buying journey in a number of ways.
[00:03:48] And we will have many leaders listening around the world today from organizations. They all collectively feel somewhat overwhelmed by the sheer volume of AI tools that are available now. So how do marketplaces help leaders move from AI experimentation to decisions that they can stand behind with confidence and prove a measurable difference, measurable impact and ROI on those tech projects too?
[00:04:15] That's a great question because I feel that not all marketplaces are the same. I think that they're also evolving, Neil. I don't want to pretend that every marketplace can offer what a marketplace should. When leaders are really looking for confidence, it's not just about more choice. And marketplaces should be helping to act to curate things and to give some signals.
[00:04:41] That means marketplaces or strong ones shouldn't just be listing tools. They should provide some benchmarks, peer usage patterns, reference architectures, and also some increasingly important assurance signals around security, regulatory environment, and scalability. That's going to allow executives to move from let's try everything to this is a choice I can defend on my board with my regulator and with my customers.
[00:05:10] This is why I went that way. I think that's really important. And I can say that security, regulatory alignment, and scalability are even more important than anything else. And I'm glad you mentioned security there. And two other areas we've got to mention, of course, is trust and governance, which are becoming rising concerns as AI becomes more embedded inside operations. So for what you're seeing with your clients there,
[00:05:39] where do organizations often underestimate some of the risks around integration and accountability, especially when we've started to talk about launching thousands of agents, et cetera? I think that's a great question because I feel that many organizations have blind spots. And the biggest one, in my view, is around integration risk, not model risk.
[00:06:02] I find organizations are spending a lot of time evaluating whether a model is ethical or accurate. But honestly, they're not spending enough time talking about what happens when the model is embedded into workflows, decisions, or accountability structures. So questions like, you know, who owns the outcome when we enact the AI recommendations? How are the decisions logged and auditable?
[00:06:32] And what happens when the vendor systems interact? This is where governance often breaks down. It's kind of about that seam between technology and responsibility, where companies are too focused on the model, but not enough focus on the integration risk. And that's where I see more and more need for us to put an emphasis, especially as they're looking to scale.
[00:06:58] And you have somewhat of a unique vantage point here, because at EY, you work across hyperscalers, AI-native platforms, and industry specialists, not to mention clients all around the world. So are you able to share an example of how alliances have helped scale an AI solution in a single vendor that they couldn't do on their own? And I appreciate, I don't want you to mention any names here, because you're going to be limited to what you can share.
[00:07:24] But any stories or use cases that would just bring to life what we're talking about here? Yeah, I can give an example. I truly believe it sounds kind of funny, Neil, but I really believe that no single vendor can cover the full AI value chain, from data modernization all the way to model deployment, especially when we talk about industry-specific use cases. I really believe the world is evolving into much more of an ecosystem play across the board.
[00:07:52] And maybe I can give one example where we have utilized an ecosystem made from a hyperscaler, an AI-native platform, and an industry-specific software. That example relates to the insurance industry, where we were looking to really radicalize cost takeout, but also improve the workflow.
[00:08:17] And we've been able to do that by combining those three providers together to really make an awesome solution in the insurance space that allows those three to operate, bringing different perspectives, but driving an outcome-based approach for the client. And that's where I see the world moving. For me, the single-vendor approach is a model that's really, I want to say, so early 2000s.
[00:08:46] It's definitely not 2026. And when we say phrases like ecosystem coordination on a tech podcast, it does sound incredibly attractive in theory, but there'll be many leaders listening thinking, yeah, that's great, but it is much more difficult in practice, or it might be from their experience. So what is it that separates marketplaces that genuinely create value from those that simply aggregate technology without solving real problems?
[00:09:13] Because I think in the last three years, we've seen a lot of tech first problem, second approaches, and we saw what happened there with the lack of ROI and lack of scaling into production, et cetera. But what are you seeing here? I totally, totally agree with you with what we've seen in the past. And I mentioned it a little bit when I talked about not every marketplace is the same. I would look at those that really create value are not talking about products, but talking about problems.
[00:09:42] What problems are they solving? What outcomes are they designed to drive instead of just a basic product push? And I think the difference here is in the intent and design. If we can look at marketplaces that are built around specific business outcomes, whether that's fraud reduction, for example, or supply chain resilience, or customer experience, this isn't just a basic technology aggregation.
[00:10:10] It's really looking to drive for a specific purpose. And in doing so, they've embedded standards, integration patterns, and governance from day one to be able to solve that real-world business problem. If it's just about a tool or a product, you're going to get a lot of product push and not a lot of outcome driven. And I think that's a real differentiator as to how marketplaces are going to more and more evolve in that direction. It really is.
[00:10:40] And you've led large-scale growth across Europe, Middle East, and indeed Africa, and worked incredibly close with European clients too. And if you look at such a vast array of data, and as you look at all these clients that you're working with around the world, how is the EU AI Act? How's that changing the way businesses think about AI deployment, especially when operating across borders? Because it's quite a hot topic at the moment.
[00:11:08] There's a few arguments that the US don't do it enough, and we do it too much in Europe. But what are you seeing? Is it slowing things down, speeding things up? Is it to be afraid of or not afraid of? Well, I'm not sure it's to be afraid of. I think it's something we need to take into account. Yeah. I really see it asking for companies to move and shift from really innovation first to almost responsibility by design. And that requires a little bit of a different approach.
[00:11:37] And you see that with the ocean in between. The US is really strongly pushing for strong innovation first, where the Europeans with the EU AI Act are pushing for responsibility by design. And that means, effectively, can the AI system be classified consistently across jurisdictions? Do we have traceability and documentation at scale? And is there governance attached to that solution?
[00:12:07] Marketplaces that really align with that are really strategic enablers, effectively, that can help on that risk classification, transparency, and audibility. I think that's the difference. It's a difference in approach from day one in terms of how people are looking at AI. I don't think we can argue whether one is better or worse. It's just a different perspective. It's definitely not something to be afraid of, though. 100% with you.
[00:12:34] And I think that push-pull between regulation and innovation, it's always been there. But it feels to me that, at least with the European approach, we've learned from the lessons of the last decade of moving fast and breaking things, and a slightly more cautious approach. And that's got to be a good thing, hasn't it? That's what I think as well. How can you argue that you shouldn't be responsible with your design from day one? It's hard to argue with that. So I totally agree with you.
[00:13:01] And for executives listening, wherever they are in the world, and they're challenged with wanting to move faster with AI but remain accountable, how can marketplaces support that better decision-making without adding just another layer of complexity and analysis paralysis, shall we say? What's the secret there? You know what? I think the best marketplaces, they can remove what I call invisible work,
[00:13:28] work the teams are doing to drive vendor risk assessments, standardized controls, and really look at reusable governance patterns. For leaders, that should mean fewer bespoke decisions and more good practice from an institutional perspective. I think they can provide better governance. In fact, speed doesn't come from bypassing governance. It comes from industrializing it. And that's where I think marketplaces can really help.
[00:13:59] It should help us get faster in how we deploy. And as we look ahead, how do you see marketplaces continuing to shape the next phase of enterprise AI, particularly in helping organizations turn collaboration into sustained business outcomes rather than just those one-off projects? Because I think those days of digital transformation being just a destination, they're long gone. It is a continuous journey. We're seeing the same with AI. But how do you see this evolving looking ahead?
[00:14:29] Well, I think marketplaces are going to change. I think they've been evolving. I don't think they're at the end of their evolution. Marketplaces today, many people would think of them almost like an app store. I think they're going to change from that. I think they're going to become more of an operational exchange, a control plane, where you bring together technology, governance, talent, and ecosystem partners.
[00:14:56] I think that's going to be big importance that you're going to see those marketplaces bring together not only those components of technology governance and talent, but you're also going to see them bring models, agents, data products, and guardrails together. And they're going to be assembled more and more towards business outcomes. That's where I think the evolution is going. Neil, I've got to be honest with you.
[00:15:24] I don't think we're at the end state yet. I think we're starting to see some of this evolve. I think it's important for the executives listening on this call to really understand how they could leverage those marketplaces and be choosy about the marketplace they choose. How are those guardrails already established? Are they really bringing more outcome-driven? Or is it just a product push in an app store? And that's what people should be evaluating.
[00:15:53] And I'm curious, you are someone that is very fortunate to have conversations with so many different voices in so many different countries, in enterprises, across so many different industries. If you were to put all those conversations into a big melting pot right now, is there any key trends or themes that everybody's talking about or asking for help with? Hmm. Hmm. I think it would come down to saying that success in AI, and if we define success as scalability,
[00:16:21] Neil, I would say it's less about the technical, logical brilliance. It's more about clarity, orchestration, and trust. And marketplaces can help leaders scale those attributes across the organization. I love that. And of course, I think one thing that everybody struggles with is the pace of technological change right now. It's moving at breakneck speed, but the likelihood is it won't move this slow again. And we're all in this pressure of continuous learning.
[00:16:51] You, in particular, I would imagine speaking with so many different voices around the world. So I've got to ask, as someone that's helping lead the way here, how do you self-educate? How do you keep your curiosity alive? And what helps you stay open-minded? Well, first of all, I'm a naturally curious person, which helps. But Neil, I have a very old-fashioned recipe, and you're probably going to laugh when you hear it. I read. I read a lot. And I read everything. And I love that.
[00:17:21] Sound bites and reading are so important to keep ahead, whether that's listening to a podcast or reading everything from various daily inputs from Wall Street Journal, New York Times economist. It really helps us stay connected and stay attached to things that are going on in the world. I'm sure you're exactly the same way. A hundred percent. Oh, I love this. And one of the questions I always ask my guests as well is we have an Amazon wishlist for books
[00:17:49] that I ask my guests to recommend and leave something for other people to read, or also a song to add to a Spotify playlist. Guilty pleasures are allowed, or it could be a walk-on song or a song that means something to you. I don't mind what, but what would you like to leave everyone listening with today and why? Oh my gosh, that's a great question. So I would leave my guests with the I Want to Break Free song by Queen because I think
[00:18:17] it's about a timeless, timeless, inspiring song for breaking constraints and driving personal growth. What a great song. I will get that added straight to our Spotify playlist. And for everybody listening wants to dig a little bit deeper on a lot of the topics we discussed today together. Where would you like to point them if they want to carry on the conversation or just keep up to speed with the developments there? Oh, Neil, I'm on LinkedIn.
[00:18:46] They can view any of these topics on our wonderful ey.com site and they should feel free to reach out. Oh, brilliant. Well, I will add the links to everything you mentioned there. And we covered a lot in a short amount of time today from how marketplaces are reducing friction and accelerating innovation in AI adoption and the critical role that they also play in enabling ecosystem coordination. So many big discussion points there.
[00:19:13] But one of the things that I particularly enjoyed was those real world examples of scaling AI solutions across industries. Love talking with you today and even had a chance to leave us with a great song too. But thanks again for joining me. It was great to be here, Neil. Thank you. I think if there's one clear takeaway from the conversation today, it's that scaling AI is an ecosystem challenge, not a single vendor decision.
[00:19:38] And Julie laid out why marketplaces are increasingly becoming a critical layer in AI adoption, especially in helping organizations connect technology, governance and operating models in ways that boards and regulators can stand behind. And you heard today why integration risk often causes more damage than model risk and how responsibility now needs to be embedded by design and why outcome driven ecosystems are slowly replacing standalone solutions.
[00:20:07] So as AI becomes more embedded in our everyday workflows, the role of marketplaces will continue to grow. But not as simple app stores, but as coordination layers that bring together tools, standards and accountability. And I think it's this shift that has major implications for how leaders plan, govern and scale AI across borders and industries.
[00:20:35] But I'm curious if this episode sparked any ideas or raise new questions about your own AI strategy, please let me know. And as AI adoption accelerates, is your organization treating marketplaces as a buying channel or strategic foundation for trust, scale and meaningful outcomes? As always, techtalksnetwork.com. You can also connect with me on socials at Neil C. Hughes. Whether you leave me an audio message or a written message, let me know.
[00:21:04] I'll read it out on the podcast. And if you've got a story to share, equally, let me know. We'll try and get you on here too. But that's it for today. So thank you for listening as always. And I'll be back bright and early tomorrow with another guest. So thank you to Julie for coming on here. Even bigger thank you to you for listening and getting to the end of the episode. And I will rejoin you again tomorrow. Bye for now.