Product Agility

Rich Mironov: The Real Cost of AI — Valuations, Bottlenecks & Business Cases - Productized 2025 TalkInTen

• Ben Maynard, Barbara Fazeka, Rich Mironov

Share episode

Send us a text

We're honoured to partner with the Productized conference in Lisbon, Portugal — an amazing conference for gathering for product leaders, designers and innovators. Productized consistently delivers thoughtful programming, practical takeaways and a warm, curious community. We're proud to be back for the third year running and grateful to Bobcats Coding for powering this Lisbon series.

This short, sharp episode features Rich Mironov on what’s next for AI, why current valuations look fragile, where the new bottlenecks will appear, and a brutally useful reminder: learn to speak the language of money.

Key topics discussed

  • Why the current AI funding frenzy looks like a bubble and what that means for product teams.
  • The real costs of AI — energy, data, and sustainable revenue models.
  • New bottlenecks that emerge if engineering friction disappears.
  • How product people should talk to go-to-market teams: make it a financial story.
  • Practical mentoring tips for product leaders working with sales and execs.

Guest bio:

Rich Mironov — veteran product leader, speaker and author known for candid guidance on product strategy, go-to-market thinking and product org design. Rich mentors product officers, speaks globally about business cases and is respected for blunt, practical advice that helps product teams connect technology to business value.



Listen, learn, act: If you build or lead product teams, this is a timely primer on aligning product roadmaps with real economics, and a reality check on AI hype.

Thanks to our sponsors Bobcats Coding for making our Lisbon series possible. Download their AI economics guidebook at bobcatscoding.com and support the shows that bring these conversations to life.

Host Bio

Ben is a seasoned expert in product agility coaching, unleashing the potential of people and products. With over a decade of experience, his focus now is product-led growth & agility in organisations of all sizes.

Stay up-to-date with us on our social media📱!

Ben Maynard

🔗 https://www.linkedin.com/in/benmaynard-sheev/

🐦 https://x.com/BenWMaynard

💻 https://sheev.co.uk/

Product Agility Podcast

🔗 https://www.linkedin.com/company/productagilitypod/

💻 https://productagilitypod.co.uk/

🖇️ https://linktr.ee/productagility


Listen & Share On Spotify & iTunes


Want to come on the podcast?

Want to be a guest or have a guest request? Let us know here https://bit.ly/49osN80

Welcome to the Product Agility Podcast where we explore the ever changing world of product leadership and org design, helping you navigate complexity and build better outcomes for your people and your customers. This week we're coming to you live from Lisbon for the third year in a row at the Productize conference where I'm grabbing 10 minute conversations with product thinkers, leaders and innovators from around the world. These quick fire chats are all about what's shaping our industry right now, from AI and product strategy to the human side of building great products. Now a huge thank you goes out to Bobcatz Coding for making this Lisbon series possible. Bobcats is a Budapest and Lisbon based digital product studio specializing in AI engineering and end to end digital product development. They're also on a mission to educate the market, exploring a new topic every six months and this fall is no exception. Their latest AI economics guidebook is out now and you can download it for free@bobcatscoding.com now here's your Talking 10. We're joined. We've got Rich Mironov, who I met last year, was it last year. Was last year. But you probably don't remember. You meet so many people because you're like a famous person, you've written books, you're well known, you live, you live in an empty conference center in Portugal. Joke, joke. Joking aside, it's so brilliant to have you here. Thanks so much. I'm thrilled to be back. Yeah. And it's. Yeah. Your name precedes you. I think talking to you last year was the first time I'd ever got to meet you and I think it was just one of such a memory for me just to kind of to meet you. I mean there's some research, more of a research about you afterwards. So yeah, it's just such a treat, such a treat to have you here and I think for everybody here at Productise to have you speaking tomorrow on business cases, our stories about money, which is just, you know, it's a great, great topic given bluntly. Indeed, I try to be as blunt as I can. I get paid to tell the truth, which is a great luxury. Yeah. How nice must that be? I'm enjoying with Barbara, I'm gonna say I'm quite enjoying saying the co. Host, co founder and CEO of Bobcat's coding. Welcome again, Barbara. Thank you for having me. And I was waiting for this chat so much like Rich is my hero in many ways in this community. So like I can't wait to start. So you want to ask the first question? Definitely. You said something yesterday at the Leadership Day that I want to bring here for the listeners. You said that now everybody thinks that product development will become cheaper because of like AI is taking over roles. But you said when the bubble of like high AI startup valuation will collapse, and you said it will collapse sooner or later we will get to know that. How much does it cost if we have to pay the real price? And could you like talk a little bit about this? Because that was the most fabulous idea from yesterday. And just to set the stage for that, my prediction from yesterday was that we're 12 months away. I don't know if we're 12 months away, but I've lived through four or five major tech bubbles, including the dot com days, and this looks like a bigger bubble than I've ever seen in my career. So hold on, it's coming. We can all smell it. And there's not enough revenue in the universe to justify the valuations of all the AI companies that have been funded. So I know it's coming. And we're going to be surprised to discover that the cost of AI is really high for three reasons. One is just the raw energy costs. I mean, all these folks building data centers so they can support their AI, somehow that's got to be paid for. The second is that almost all of it's being given away today, so people are using it because it's free or cheap. Somehow we're going to have to generate a tremendous amount of revenue to support those companies. And then the third thing I think is going to bite us is the AI companies have pretty much consumed everything that's ever been written in the history of humanity to fuel their models. I don't know where we're going to find another set of data like that to let them grow next year. So we've chewed up and used almost everything that can be borrowed or stolen or lifted or, you know, whatever for these models. The idea that we're going to have all the AI models generate their own data I think is fallacious. So, you know, we're running up against a wall where the AI companies I don't think can support themselves, can keep the data going, can avoid raising prices. And so we as consumers of this are going to get slammed with some really big bills and or the tools we depend on are going to instantly vanish and go away. So it's a particular smelly bubble. Picking up what you said, it doesn't smell right. It doesn't smell right. And I think the economy, the economics are clear. There's a Great Harvard economist named Kenneth Rogoff, who wrote a book called this Time It Will Be Different. And he has like four or five hundred years worth of bubbles, the South Seas bubble and the Tulip bubble and all these things. And there's always a group of people who have a vested interest in standing up and saying, this time it's going to be different. There is in fact infinite capacity to use whatever it is that we're doing. We're not going to run off the edge. But yeah, the mortgage back crisis, it always ends up so we've seen it before, we'll live through it, it'll be fine. But you know, look for that really big deflation coming up soon. Bubble is people getting overexcited and things becoming overinflated beyond its actual true value. And then the bubble pops and we're left with, we're left with something. Sometimes we're left with very little. Mortgage backed is a great example. But other bubbles were left with something. What do you think could be left? So there's a lot of goodness in this whole AI wave. I think the idea that we could build software 20 or 30% more efficiently is probably right, maybe even 40% efficiently. But we're not going to replace all our product managers and we're not going to replace all our engineers and our designers. That's just never been true. So there'll be tools, companies that survive, there'll be companies that put these into their products. That'll be great. But the trillion dollar valuations on companies that don't actually have any revenue, investors are going. Warren Buffett's favorite thing, right? When the tide goes out, we find out who's wearing their bathing suits. Right. Some of these will survive as they have in every boom and bust before, and we'll see where the value is and the lucky people are. I have a question connected to this. We met in June at a productized meetup here in Lisbon and we were talking about what's going to happen when we will have wipe coding integrated into our everyday life and everybody will be able to push out like 12 features while having their espresso. And you said that there will be a new bottleneck in the system. And I guess that's what we're sharing with our audience today. Sure, yeah. The obvious thing. And anybody who's done systems design knows that there's a bottleneck in the system, which is the slow step. And when we fix that bottleneck, there's a new bottleneck in the system, and that's the slow step. And when we fix it, etc. So imagine a world where we can spin up hundreds of millions of new software products every day, and then we're at the receiving end as consumers and buyers trying to figure out how we're going to sort through all those things, or as marketers and salespeople, how we're going to sell them. If we completely remove the engineering step from the bottleneck, we've still got to figure out how to price and package and sell and market and support things. And I don't see companies, at least in their current structure, figuring out how to sell 100 times more products at 80 times the Velocity and not, you know, and not tripping over their own shoelaces. You said with the current structure, do you see a world where there is an evolved, a different structure that, you. Know, somebody will propose it and it'll either work or it won't? I don't know what it is. If I did, I'd be investing in it. But it's funny, I was talking about predictions earlier and I was thinking about what we saw in the Agile community over the last 20 years and the game was always up. When you made, when people have made their money from the base level understanding, when it's like team coaching. It wasn't team coaching, it was enterprise coaching. When it was like business agility, mistakes were raised and raised and raised and it popped. And there's very little left from that, from that money making part of it. I was wondering with the AI bubble, how long is it before we see AI for organizing organization designs and we're going to start up in the game? I think there's a lot of really good applications for AI in a lot of spaces and we should apply them. I worked on some AI in 1980, it didn't work yet. I worked on some AI in 1990 and it didn't work yet. This is not a new thing. But the tech is bigger and the data sets are bigger and the spending is bigger and the venture capital is bigger and it's real. Now, you know, do I want some AI systems redesigning my organizational structure? Probably not. I don't think I want it evaluating all of my employees. But I'm perfectly willing to have it write all the tests for my system that I then have real test folks review. Probably don't want to do all my cyber analysis because whoever's putting stuff into the system may in fact be seeding it with things I don't want it to seed, but, you know, use for purpose. Do you enjoy AI as a topic or do you find it? I'm tired of it. Yeah. We've been having the same conversation for a year. Every meeting, you know, every CEO and every conference is how are we going to make AI the thing we do? You know, I remember when the Internet was born and I remember when mobile devices were born. And I remember when, you know, first, second, third, fourth and fifth generation languages were born. And those were all good things. They didn't actually change the world. So, you know, we can rerun all of our Arnold Schwarzenegger movies, but I'm not ready to see sign up for Terminator yet. No Running Man I could go for, but you know, it's a great film. No, I just kind of sense that this is a. It's a tiring topic and we go round and round and it feels that no one's really pushing that much forward. We're almost waiting for, like we kind of. We've reached this point where we're almost waiting for the next, next massive leap. But you were saying, like, there's no more data to train it on. They said, what do we do? What's your advice, Rich? If I look at the last two or three or four big trends, right? And again, I think of browsers and the Internet, 1995, it took us five years or eight years or 10 years to really figure out what that stuff was good for and what it wasn't good for. And after the first year or two, we didn't have Internet product managers. We had product managers who thought about where the Internet stuff fit in their products, right? In 2008, the iPhone was actually not a hit yet, but itunes and the App Store turned it all around because suddenly people could build mobile applications and they didn't have to be a phone company, right? A mobile device company. And almost all the mobile device companies went away because they only had their 20 apps and the App Store had 1,000 and then 100,000 and then 5 million apps, right? And every board of directors in the world was beating up on their CEO to ask, do we have an app in the App Store? But nobody actually knew what it was for, right? And it took us two years or five years or eight years to figure all out what things were going to be. Apps and not and what was going to work and E commerce and all this stuff. The adoption cycle has to happen and most of the tools aren't the right tools and most of the ideas aren't the right ideas. But we bumble around and run into the walls and spill a Lot of venture money on the floor and punish ourselves and learn something. And I think we're in that bumping around moment where everybody's got an idea. Most of the CEOs I talk to believe that we're going to get a 50x improvement in engineering and we don't need product management for that. We'll just. We'll just connect whatever the sales team wants to post it notes to automatic co generating and we'll replace everybody and we'll just ship everything all customers want. It's a horrible idea, but that's okay. And it'll take a while for us to connect the reasonable improvements with reasonable expectations. And that's going to be disappointing to a lot of people. People. But it's progress. And you know, five years from now, we won't have AI product managers. We'll have product managers who know where AI helps them in their jobs and where AI helps their products succeed. And it'll be in the toolkit and we'll all know what it is, more or less, and we'll move on to the next bubble. I do wonder that the shift that I'm looking forward to seeing is when AI comes after the word rather than before. Because I think that's. If we can see that, then I think people are thinking more sensibly about. But we are overtime. Rich, thanks for bringing us back down to ground. Sure. Good. Did you guys want to talk about business cases or stories about money? We forgot about that, didn't we? We can go over a little bit. Let's talk a bit longer. So your talk. I think so many lessons led us astray. Barbara, I'm just trying to keep you on top of it. No, you're in the wrong seat, Rich. That's what I'm learning today. You're in the wrong seat, man. Do you want to interview yourself, Rich? What question shall I ask you next? So if I were going to summarize the talk very quickly. Go on. Yes, please. I got too into it. You're supposed to ask me to do that. Oh, shit. Could you just for the listeners, would you mind just giving us a nice summary of your talk? And when you say business cases, stories about money, what do you mean? It's actually a very simple idea which almost no product folks get, which is that we love what we do and we're fascinated by engineering processes and product operating models and backlogs and sprints and all this stuff. And it turns out that almost none of our business side executives and partners care at all about that. And so we're the guy at the cocktail party who's insisting on spending a half hour explaining golf or Dungeons and Dragons or whatever we're obsessed by to people who couldn't give a rat and just want to figure out how to escape and chew off their leg and get out of the room. And so every time we launch into a half hour dissertation on how we build stuff, our partners across the way in marketing sales hear that they're not going to get what they want and that we're wasting their time and they don't need us. And so the language of money is where we put short sentences together that have currency symbols in them and we say if we could build this platform for you, it might be worth 10 to 30 million dollars. Is that interesting? And if it's not, we move on. And if it is, we work on it. But my claim, and then we'll let it go is our go to market partners and go to market execs can't hear anything we say if the sentence doesn't have a currency symbol in it. And so we're wasting our time and frustrating them by having these long discussions, discussions about how we do stuff and how much we love it and discovery and, and on and on when it's the wrong audience entirely for that kind of conversation. Yes. Don't really give a. And it comes down to we make. We said to Chris Compston on earlier, he says about making money, saving money or protecting money. And yeah, generally speaking, it seems that's what everyone cares about apart from product and engineering. A lot of the time we're fascinated by, you know, which process we're running in. Is it Scrum or Kanban and nobody. Else cares and they shouldn't. Was that sufficient Rich? Works for me. Yeah. Did you want to ask a little question? Go on. Yeah, be quick. Quick, quick question. You are mentoring a lot of product people and like, how do you help them to get over this menacing, like psychologically, like not hating salespeople from the course so they can like be connected. Yeah. All I do these days is coach two product officers and, and two things. One is we have to appreciate what they do and who they are and they're not us. Right. So they're money motivated and they're short term thinkers and they're really good at selling and that's what brings money into the company. And if they were engineers, they'd fail at their jobs. So what we have to do is we have to understand their behaviors on average and expect them to escalate to the CEO every time we say they can't have what they want and to work around us so that they can close their quota and pay for their yachts. Right. That's who they are, that's what they do. And so rather than resenting them, we have to understand their motivations and where they sit in the organizations and their jobs so that we can do our job and we can help them do their job and champion the things they're going to do that are useful instead of being resentful that they make three times what we make. I love it. And on that bombshell, it felt like a bombshell. Shreem's sick of me saying that. We'll draw it to a close. Rich, mate, it's a pleasure. It's always good. It's the end of the day and you're really engaging, so you're nice and easy to listen to. It's topic we both really care about. So thank you for coming along. It's my pleasure. And Barbara, thank you as always. Great co host. We'll be back again at some point soon. We're just about to record our last episode of today actually. So yeah, it's been brilliant. Come back again soon. Thank you very much for being here, guys.



Head Shot

Ben Maynard

Host