Latest / Fit Happens: Why CEOs and Leaders Succeed or Fail / The Boardroom Is Broken: How AI Changes Executive Leadership Forever

Transcript

The Boardroom Is Broken: How AI Changes Executive Leadership Forever · Fit Happens: Why CEOs and Leaders Succeed or Fail

• Jason Baumgarten [0:00:00] I'm Jason Baumgarten and you're listening to Fit Happens, the podcast where top leaders, investors and board directors share the stories, surprises and hard earned lessons behind finding the right fit. Let's get into it. Welcome to the podcast. We are on Fit Happens where we explore the leadership reality that fit — when a job, company and team clicks — is when we're at our best. I am delighted to be here with my good friend Steven. Welcome to the podcast. Steven Wolfe Pereira [0:00:33] Thank you for having me, Jason, and love the title of the pod because Fit does happen. Jason Baumgarten [0:00:38] Listen, we've got to start with who are you? Tell everybody a little bit. Steven Wolfe Pereira [0:00:43] Who am I? I like to think of myself as a father first, a good friend and husband. But professionally I think I've always been an entrepreneur at heart, just always fascinated by technology. And so even though I started my career very early in finance, I was always in kind of technology, media, telecom. And that has really been the through line throughout, you know, a career almost 30 years now. But I really view myself now as an entrepreneur, technologist, board director, and really trying to be a champion for what I believe is going to be one of the most important things in this new era which is around AI governance. Jason Baumgarten [0:01:25] And how are you doing that? Steven Wolfe Pereira [0:01:27] Well, funny you should ask because I started my career in tech at this little startup called Akamai Technologies way back when in 1999, started in their network infrastructure engineering group. And pretty much since then I've been dealing with data and machine learning. So this whole idea of AI back when it was ML, I'm sure you remember those days, but we've been doing this for a minute and when we had the ChatGPT moment in November of '22, it was just like, wow, finally it's here. We've been talking about it for a long time, but it's really here. And I thought, where could I actually add value? How could I have impact? And it kind of dawned on me the most antiquated part of the enterprise is actually the boardroom. And so I was like, what can we do to kind of super serve and really help directors and executives? Because you need both in the boardroom. How can we help them get ready for — again, what I believe will be the most important thing — which is going to be the governance of everything that is agentic, everything that is coming with AI. And that's how we started Alpha. And so what grew as kind of this idea to provide AI governance intelligence to the boardroom has now kind of evolved to be this multifaceted business where we have at the heart of what we do, the Alpha Institute for AI Governance, that's our independent research arm, also doing certification for the boardroom, as well as governance ratings on both tech companies and boards. And then we have part of Alpha is a really great global leadership network, kind of a peer group. Think of it as like YPO for the boardroom. So that's the Alpha Council. We also have a boutique advisory business. I didn't want to be in consulting, but we get so many asks from boards to kind of help them understand how to navigate this. And because we're independent, you know, we're able to call balls and strikes. So the difference is some of our council members have become partners, and so it's really going to a boardroom as not just operators and technologists, but also as directors. So kind of for directors by directors. And then we have our content arm. We put on, you know, really substantive events with partners like NASDAQ and the New York Stock Exchange. And then we have our own platform for publishing, as well as a newsletter and podcast called Board Brief. Jason Baumgarten [0:03:33] Well, having been to a number of the events over the years, I will say two things. One is you are one of the best connectors in the world, so you always have wonderful people. And two, you really foster real conversations. And I think that, you know, it doesn't feel like nobody's got the agenda of the brochure where it's always like, hey, what's really going on? And in a space that's moving so fast, it's wonderful to have that kind of community. So I'm thrilled with what you're doing with Alpha, but just to that — Steven Wolfe Pereira [0:03:59] I think it's really important. We're at such a point where there's so many charlatans that are out there that are all of a sudden AI experts. You see it all over LinkedIn. And haven't you seen my tattoo? You know, AI expert, right? Jason Baumgarten [0:04:13] I had this before '22. Yeah, that's right. Steven Wolfe Pereira [0:04:16] You're an OG. But it's a real issue, right? Because all of a sudden everyone's an expert. And certainly we're in now the gold rush. And we kind of grew up in an era where we saw this during the dot com days and we saw all the charlatans there. And so this is not a drill. And because this is going to have such monumental societal, geopolitical and economic effects, I think it's really important that we actually have substance. And that really is the spirit that has kind of guided us — because technology is the North Star for us, but when we can be in a room, there's no room for bullshit. And I think that is really kind of the litmus test: can we actually not hear from a consultant, not hear from the charlatans. Let's actually hear from the builders, from the operators, the people that are actually building the companies, leading these initiatives, and then to the folks that have to make these decisions that are in the boardroom — again, the directors and the execs that are actually having to manage all of this. Jason Baumgarten [0:05:12] It is such a difference when you talk to people who are really in the thick of it, building stuff with this technology, versus folks who are in very controlled corporate environments — they might have one tool that's unlocked in a very limited way and they really aren't seeing the latest and greatest. I was sitting down with somebody at one of the big AI companies recently and they said, let me show you what's coming. I have to say, the coming attractions almost made me want to put down the tool for a month and wait, because I'm like, it's going to get so much better. Steven Wolfe Pereira [0:05:42] Yeah. If you're using Copilot, no disrespect to Microsoft, that's not using AI. Jason Baumgarten [0:05:47] There's great stuff coming, though. It is changing. It's amazing. But I want to go back because you went quickly over your career. You've had so many amazing jobs, chapters from being an entrepreneur to being at some of the most incredible startups that then got bought by big companies. You've been grinded out as a founder. Is there one role? And I don't even want to say job, because it could have been like, no, no, no, there was a year. It was a year that was special. Is there a role that you had when you think back on your whole career and go, man, that was the moment where I was in flow, where I could work, you know, 100 hours a week and didn't feel like it. Where I was just at my best. We'd all love to say that's the one we're in now, but sometimes not. Sometimes it's something we did early in our career. What's that one for you that really stands out? Steven Wolfe Pereira [0:06:33] I appreciate that question. As you know, I'm a huge football fan — or soccer in the US — but, you know, kind of football is life, as we like to say. And even though I kind of sucked as an athlete, I think it skips the generation — my son's really good. I like to think of myself as a professional business athlete and I think the whole idea of flow is a really great concept. And I distinctly think of three times in my career that I was really in flow. And I would say the first is when I kind of left finance and kind of private equity to join Akamai. And those first few years, I was just in flow. It was the most incredible period to be building such a foundational company for Internet infrastructure at the time. And, you know, I was in my mid to late twenties and what did I know? But all I did know was there were 60 plus PhDs from MIT and Carnegie Mellon, all these other places, and I was the dumbest person in the room. But I was just learning. And we were literally building the Internet, you know, brick by brick, right before our eyes. And so that was just such a fascinating time. We went through incredible highs. I was there pre-IPO, we took it public. Incredible lows. You know, during 9/11, we lost our founder and CTO and my dear friend Danny Lewin, who was the first person killed in 9/11. Fun fact that probably people don't realize: the technology that he helped co-invent literally saved the Internet that day. Because if you were not using Akamai's technology, your website was down. You literally could not get information. So it just goes to show the impact that tech really has in the world. But those first few years where I got to really touch every area of the business from engineering to finance — I worked for the CEO at the time, George Conrades, helping run everything from M&A, corporate finance, investor relations. So it was just such incredible talent that was there. And you see kind of this diaspora of Akamai alumni that has since left, and you see them now kind of popping up 25 years later, now leading the AI revolution. You know, for example, Arvind Jain, who's the CEO of Glean, which is, you know, just an incredible work AI company. They're absolutely crushing it. You know, he was one of the engineers at Akamai back in the day. So I feel like that was one time I was in flow. I would say the second time was really kind of my first true C-suite role, which was when I got tapped to become the Chief Marketing Officer for a company called Datalogix. Probably one of the hardest transitions, because when you go into the C-suite really for the first time, there's so much that you don't know. And I learned so much from the CEO at the time, Eric Roza. And we were going to go public, but we ended up getting acquired by Oracle for over a billion. But that was just an incredible time, an incredible experience, and again, using data and machine learning to really kind of figure out how to connect online ads to offline sales. So we had purchase data, and it was very exciting. So really creating a new category. And then candidly right now, building Alpha. I've never been so energized to build. And, you know, you hear Andreessen Horowitz talk about it all the time: it's time to build. And I believe deeply in that because I think the biggest competitive advantage that you could actually have right now is speed, because the technology is moving so fast. The curve that we're on is not linear, it's exponential. And what an incredible time to be building with all these agentic tools. It's like you're a kid in the candy store. So you have just such access to incredible intelligence that we've never had before. So what a great time to kind of build something that is truly AI native from the ground up. I feel like I'm in flow right now. Jason Baumgarten [0:10:14] I mean, I find so many people barely have one — so to have three is a special place to be. There's always that chapter, that role, that job that was like, man, that was perfect on paper, but just the opposite. Is there a moment where you're like, yeah, that's the one that stands out, as — you know, I got it wrong. It looked good, wasn't me. Steven Wolfe Pereira [0:10:35] You know, I think part of my philosophy — we'll come to it towards the end — but I'm kind of a learn-it-all kind of guy versus like a know-it-all. And I think every experience that you have, it really depends on your worldview, right? Like, oh, you went to a job and it was a failure. No, I think I just learned lots of things. And learning what not to do or what you don't like, I think, is equally as important as knowing what you do like. And so I think one of the roles, oddly enough, it was for an AI behavioral analytics company called Quantcast. You know, I went in and I was, you know, kind of off of the Datalogix role, then kind of CMO over at Oracle Data Cloud and then Neustar. And now I got tapped to be the Chief Marketing Communications Officer at Quantcast. And I just think that at that company, they had incredible talent. But the fit in terms of the marketplace with the technology — and maybe it was too early. I mean, literally, this is like 2017, 2018, 2019. I'm talking about AI and just it would fall on deaf ears. Like, it was so early. And so to kind of really talk about something, you know, where you could see where the puck is going to go, but timing really is everything. And you always hear about product market fit. I feel like it was probably just a little bit too early, but I made some incredible friends. There were a lot of Brits there, incredible mates. And I got to work with some incredible talent that in fact are today still involved with Alpha. Jason Baumgarten [0:12:00] I mean, I love talking about, like, what is somebody's superpower? What is the thing they're uniquely good at? And interestingly, most people are not aware of it. It turns out less than 15% of executives actually have the self-awareness. It's not because they're bad executives, but the things they think they're good at are not actually the things they're good at. The thing that they're uniquely good at, they take for granted. They skip over it. They're like, yeah, that's easy, I don't have to talk about that. Look at my financial modeling skills. You're like, no, no, no. But you're actually a great synthesizer. That's your real gift. You've reflected a lot. You've had this amazing career of being able to jump between big and small and fast moving and things that were ahead of their time, but maybe not ready. What have you distilled your superpower to? Steven Wolfe Pereira [0:12:42] So I think it's really important to know what you are great at and also to admit and be intellectually honest about what you're not great at. And I feel that I learned some of those skills from two specific leaders. One is Paul Sagan, who was the president and then future CEO of Akamai. I learned so much from him. He was just such a human leader — I don't know if that makes sense, but he was so real. And I just vividly remember, you know, when I kind of moved from network engineering into finance to kind of run the corporate strategy role. I was like, we're going to get you on the road, Paul. Like, let's go. We got to meet all these investors. We got to tell story. And he was like, you can't mess with my family. I do carpool. I take my kids to school in the morning. Like, it was just so interesting to see this guy who's obviously, you know, incredibly successful, such a pioneer — and he was really balanced in terms of work and family. And I really took that: oh, there is a way to do that. So I learned a lot from him. And he was one of the folks that kind of started to identify: hey, you're really good at connecting people, places and things. And then later on in my career when I worked for the CEO of Neustar, Lisa Hook, I learned so much from her and she also kind of identified that — and just really understanding that if you are good at something, it's not to just take it for granted. You should triple, quadruple down on what you are great at and then hire around you and, you know, check that ego out the door. You don't have to do everything, but you really need to know what you're great at so that you could figure out your lane. And when you think about everything from a board skills matrix or whatever, you could articulate what you are good at and what is going to be the value that you could actually bring to the table. And I think that actually becomes really key when you are a builder, when you become an entrepreneur. Jason Baumgarten [0:14:35] Yeah, it's such a good segue because I think, you know, when you're trying to do something with other people, you've got to figure out what's your piece of that puzzle and what's theirs. You have to be really honest about what you need help with and welcome it and go get the people that are excellent at other things. So it's fun to hear you tell that story of also how someone else can help you figure that out. It's not all about, you know, journaling and deep walks in the — Steven Wolfe Pereira [0:14:57] Look, I would not be where I am today if I was not standing on the shoulders of giants. And I mean that so sincerely. Like it's easy to kind of say cliché things. But I think throughout my career I was just a kid of immigrants who didn't know anything about finance or technology. I mean, I had a Commodore 64 and I tried to bang programs out on that, but I didn't have anyone teaching me how to code. I didn't have anyone. Again, think about early 80s. There was no Internet, there was no way to kind of learn these things. I wasn't surrounded by people, but I think I was always a computer scientist at heart, a technologist at heart. But back then you didn't know that. And so I found those things out kind of like in a Sims game, you know, just like you're in the dark and you're just moving in one direction or another and people help you along the way. And I didn't know anything, but for some reason people either said, hey, you know what, you've got hustle, you've got grit, you know, you're actually kind. I think it's one of these underestimated things — like, you are polite, you say thank you, you send the thank you note. Like, just simple things: you actually treat people like a human being, with respect. And I feel like those are values that I learned from my parents. And, you know, being a son of immigrants, like, you're always taught the one thing that you can never take away from you is your education. And so how do you really kind of pay that forward? How do you ask, how can I help? And I think that was always my intent when I would meet people. I'd be like, how can I help? And all those early jobs, you know, I would go through a brick wall for people. Like, I would just always try to do my best. And I wasn't the smartest. I didn't have all the pedigree, but I was always the first one in, last one out, and I would always work harder than everyone else. And so I think over time, the consistency of that pays dividends. And when people say, hey, you know what, I'm going to introduce you to this person, they're looking for an XYZ — they're looking for a CMO here, they're looking for technologists over here, they're looking for a place to kind of invest in a startup — like, those opportunities happen because you've earned the right that people would recommend you. Jason Baumgarten [0:17:05] Yeah, it's such a great story of — there are things you can do quickly in life and things you can only do slowly over time. I always tell my kids that the hardest things you had to start well before you knew you needed them. You have to dig your well before you're thirsty, as the book once said. And I think with a lot of things around relationships and investing in learning, you can't wait till you need it. You've got to do it, you know, well in advance. You also touched on this idea of, you know, everything I have, I carry with me. And I think that it's a very immigrant mentality. My grandfather used to say that all the time. You know, you'd say, why don't you buy something? He'd say, everything I have, I have on my back. It's a spirit of learning and education being such a powerful thing as well. And I think living in this Internet moment, people always ask me, what should I do? I said, learn. You know, the number one thing you can do right now is just learn, be better. Steven Wolfe Pereira [0:17:55] But think about the moment we're in, right? Like, again, I think back to my 12-year-old self, right? Jason Baumgarten [0:18:01] Playing King's Quest on an x86. I remember those days. How did I do? Did I date that? Steven Wolfe Pereira [0:18:10] Well, that's pretty impressive. But think about what people are exposed to today and it just — it blows my mind. And you know, my kids are young — my daughter's 8, my son is 10. And I think the most important thing, because if we're going into this era of just abundant intelligence, super intelligence — what do they need to do? How are they going to thrive in the world that we're entering? And I keep telling them: learn how to learn. Right? I want you to not use these technologies as a crutch and end up like WALL-E in some hover chair being entertained by a screen. No — use these things. But you need to learn how to think, consume, ingest. And this is an experiential technology. I think that's what people don't realize: you're never going to learn how to surf by reading a PowerPoint. You need to use these tools, build whatever it is, you know. And so I just think — where you go to learn, of course it's YouTube, but every single company — I mean, think about what Anthropic has put out. They put out their whole education. They have libraries, they have videos, courses, white papers. Like if you just ingest all of that, listen to all the podcasts, I mean, everything out there — there's so much to learn from. So I feel like most people are lazy, if we're being honest. All you've got to do is just look up and just start somewhere and you end up going down this virtuous rabbit hole. It'll lead from: oh, what is a token? Okay, how do I actually build an LLM? Oh, how do I actually build an agent? Like, you get to kind of go down this rabbit hole and all the tools are there to help you build. That's what's exciting. That's what was missing when we were growing up. Jason Baumgarten [0:19:57] Yeah. Well, I will say saltwater is a lot better on a video than in your mouth. But learning how to surf is a lot of pounding — AI is a lot easier. Nobody's trying to drown you along the way. But you know, this idea of this change moment we're in — in your work in governance, we both spend a lot of time with directors. I'd be curious: if we zoom back five years ago, if we were having this conversation five years ago — Steve, what's a good director? What do you think you would have said five years ago? Steven Wolfe Pereira [0:20:28] I might be the outlier here because I think that we've had a lack of what my benchmark for good is. And again I look at these things through a technology lens. And so we have been dealing with this lack of technology literacy for 25 years — call it plus or minus, whether it's data, whether it's cyber, all these issues. And now we're at this moment of AI. And so I would argue that most folks aren't technologically literate. And it really kind of took something like the Enron fiasco and Sarbanes-Oxley to kind of really say, hey, you know what, we actually need to have qualified financial experts on the audit committee. Right? I mean there weren't QFEs before that, which is insane. Right? So why is it okay that for 25 years, oh, if it's a tech thing, oh, just give it to the tech person on the board. I'm going to be, you know, just doing whatever I'm doing over here. But you know, we actually have one person that will be responsible for that — give it to that person. I don't think that's actually responsible governance. And the whole idea of the learning and development initiatives of the nominating and governance committee — I don't know if people really take it seriously or no one ever uses the budget or whatever it is. But I think what does good look like back then? It's the same as now. It's that you actually have the board that is truly engaged to be the steward — the duty of care and duty of service — understanding where is this company going to be in the next five years. And again, hire and fire a CEO, but really the corporate strategy part of it. And you cannot deny the impact that technology runs everything now. So how can you be illiterate about the technological shifts that are happening? And we've seen it over the past 25 years. Obviously we saw the Internet, we saw mobile, we saw cloud, and now we're here. And so I think that the jig is up, to be honest. Like the whole idea of noses in and fingers out and that we have our nice cushy quarterly board meeting — you know, we meet four or six times a year, we're going to just kind of casually meet and discuss some things and get some reports — I think those days are over. Because if we're going into what is now going to become the agentic enterprise and you're going to have an agent economy — I mean, literally, the CEO of Cloudflare came out saying that we're going to have more agents than people starting next year — I don't think that's a crazy prediction. So if you now have more AI agents than humans on this planet, we are entering a world of continuous everything, and that means continuous governance as well. And so I feel that if you look at where we are, the tech companies are too big to fail and they're just trying to win this AI race, so they're really not focused on governance, if we're being honest. You know, geopolitically, we're in this AI Cold War, right? So US and China — those are the only real players. And so we have to win, whatever win means. And obviously there are massive geopolitical ramifications of those decisions and the things that are happening, but there's no governance or guardrails there. So where are we going to see governance? Because this is going to have impact not just on a business — it's going to have an impact on society. And so I think the boardroom is truly the last hope we have for any type of AI governance, because they are actually fiduciaries. And when you have to make the hard decision of what are we going to do when our competitors are actually using digital labor for pennies on the dollar — 24/7 — and they are letting go half of their team or maybe even more, what are you doing as a company? Those are the decisions that are going to come to the boardroom. I think people are underestimating the very, very hard decisions that we're going to have to make over the next three to five years. Jason Baumgarten [0:24:29] I think you've pointed to a shift in the responsibility and the impact on humanity that the private sector will have that we really haven't ever seen before. I mean, the agricultural revolution was the biggest labor shift — like what we're potentially seeing — but it happened over so many decades that it was slow and gradual and very dispersed. And so there wasn't a small number of companies that were responsible because it was small farms, you know, largely mechanizing and losing labor. So I think it's going to be a very interesting five to ten years as we navigate this. And I love that you're putting the emphasis on the responsibility of good governance — not only the company, but companies need economies to survive. And so how do you think about your role as a player in the economy? I mean, somebody said to me, well, it's great, we're getting more productive. I said, well, you need people buying your stuff. Have you thought about what happens if everybody gets more productive? Steven Wolfe Pereira [0:25:23] Customer. Right. Jason Baumgarten [0:25:25] You kind of need a customer. And I'm not sure everyone's getting retrained to be, you know, mechanics and roofers. Steven Wolfe Pereira [0:25:32] Right. Jason Baumgarten [0:25:32] I think it's going to take a while. One of the things with AI — and you know, I'm digressing a little bit, but I think it's important when you think about fit, actually — is this idea of it's so good already at getting to some polished answer that what I'm hearing from a lot of executives is — and I'll use a comment from one of my college professors — you know, we used to have reading in the library. You either sat there and read it or you could pay and get it photocopied and take it with you. Steven Wolfe Pereira [0:26:01] And library was like actually a class — it was a session. Jason Baumgarten [0:26:04] Yeah, well, you had to learn how to use it too. Like, you know, Dewey Decimal system. But I still remember this professor and he said, how many of you went to the library and got the reading? And we all raised our hands. We had one guy in the corner, you know, didn't. And he said, how many of you photocopied it? And most of us raised our hands because it was a lot easier than sitting in the stacks and reading. And he said, I want to be very clear that photocopying the reading and taking it home with you is not doing the reading. And it dawned on us that a lot of us were like, yeah, yeah, we did the reading because we had it — we had that physical copy. We could now read it. And he pointed out that some of us were maybe not getting through it as quickly. And I have this moment sometimes with AI where I'm like, it's jumping to this beautiful completed answer. And a lot of executives are increasingly saying, no, no, no, you've got to break it down. You've got to do the check-in steps. You've got to nurture it through to make sure that you actually are putting the thought in between A and B — not letting it jump all the way to B. Because by then it's just the photocopying you're holding around. You don't have time to go through it. How have you been — you know, because you're such a great learner and practitioner, just take a minute and talk about, you know, how are you not letting it just get to the end and going, yep, it's done. It thought for me. Steven Wolfe Pereira [0:27:23] So I think this is such an important point on so many different levels. And this is kind of what we're seeing now — I mean, literally just go onto any social network these days, it's all slop. If I see another, you know, kind of GIF on LinkedIn where they're trying to show something and obviously it was AI generated — I mean on Pinterest, I think 90 plus percent of all images on Pinterest are now AI slop. I mean, it's insane. So you're just seeing all this slop flood the zone. And when you really look into it — are you learning, right? Because you could actually output. And this is just like anything that we've seen in digital: garbage in, garbage out. You're just doing it at scale, right? So now you're doing it at AI scale. But Ezra Klein — you know, I always love his podcast, it's very thoughtful — and he actually talks about this a lot, which is, you know, obviously he's using AI tools, he's very smart and savvy about it, but he talks about his process as a journalist and how you need to actually do the work. You need to read the source material, you need to kind of do that initial first pass — because that is the judgment layer. And the judgment is where the human is still very much in the loop. And in fact that might be the only role that humans need to really play when you think of cognitive labor, because anything done by keyboard, video, mouse — a machine can do it better. And so this is what Emad Mostaque talks about. He was the co-founder of Stability AI. He helped really develop the first text-to-image large language model. And he talks about how cognitive labor is cooked. And if you actually think of digital labor basically doing everything for pennies on the dollar and now you actually have the marginal cost of software going to zero — well, does the value of human labor go negative? And that's like a radical concept to kind of wrap your head around. But what is the value then of a human? It's going to be your judgment — and you have to do the work. Literally every single VC that I know — you name it, all the big firms — they are giddy because they are now saying, you know what, we have a completely new mousetrap in terms of the way we actually start companies. We don't need you to have a technical background, you don't need to be a developer, a technologist. We are looking for people that actually have subject matter expertise — they actually understand a niche industry deep and they have relationships. Because we will use AI to do all the tech, but we can't unpack the nuance that actually is not inside of an LLM. And so understanding the judgment — and we're moving from a world where people are still talking about prompt engineering, that was two years ago — we're now in the world of context engineering. As you well know, understanding context is critical. And then the next evolution of that is intent engineering. How do you actually take the intent of what you are trying to do and build that into an agentic workflow? And I joke — I think the buzzword for 2026 is workflow. Because now you actually need a human to understand: what are all the steps in this workflow and how do you make it agentic? And that breaks down the modern enterprise as we know it, because it doesn't care about your silos. It doesn't care if someone touches something in marketing and another person in product and another person in finance. No — what does the workflow look like? And that is going to be so radical that people are going to have to reimagine the way they do work in a world where it's going to be human and machine. And so all this is to say: we are just at the beginning of this process. But yes, I think you really need to understand human judgment — and that is going to be critical. Jason Baumgarten [0:31:39] Well, it's interesting — the other thing we're talking a lot about is this combination of judgment and prioritization. Because what's interesting is you used to have to prioritize because resources were scarce, but now you have to prioritize because output is scarce. And what I mean by that is — not can you create gobs of output, but your customer, your client, your reader, your observer, whatever your end market is, only has limited time and money. So if I have limited time and money — LinkedIn's such a good example — I know you write your posts, I write my posts, you know, it's my Sunday activity with a cup of coffee nine times out of ten. And the reality is that that prioritization of what do I want to put out there — even if I'm using Perplexity for digging up some research or whatever — the reality is that prioritization has shifted so radically from the capacity of the organization and really managing physical assets and human assets to: I now have almost unlimited, quote unquote, human assets in the form of agents. I need to do a better job of prioritization, and then, to your point, judgment on the back end. But it does speak to the role of a leader and the role of a board radically changing. Because the whole idea of why does a business exist — that was all to reduce friction. Well, we're going to enter a different world. Steven Wolfe Pereira [0:32:57] Well, I'm sure you saw the HBR article about AI doesn't reduce work, it intensifies it. And so when you have so much more volume, there's not enough hours in the day for a human to ingest this. And again, you could have an agent doing it. But now we're back full circle: how do you prioritize? And what's the judgment? Is this good work? Right? Because again, you could just put so much slop out there, but still, someone needs to make the call. And that's why I think the future organization is going to have this kind of three layers. The bottom layer is going to be your autonomous layer — so that's going to be all AI agents doing all the monotonous stuff. You pick the department — whether it's finance, legal, HR, marketing, sales, whatever it is — you will have AI agents 24/7 doing all that monotonous stuff, because anything that can be automated will be. At the top of that, you're going to have the judgment layer, and that's where the people will be. And that's going to be about what is our business strategy, what are our thoughts on being creative — right, creativity is going to be critical. Governance — right, human judgment is going to be at that top tier. But then in the middle is what we call the messy middle. And that's going to be maybe humans managing agents, or maybe it's going to be agents managing humans. And this is where all bets are off — I don't think we are ready for that world. Jason Baumgarten [0:34:19] Well, it's one of my absolute beliefs that great leadership starts with clarity. You know, what are we here to do? And how do you fit into that? And it's amazing how many people struggle with it, right? They have grand ambition, but they struggle to turn it into clarity for the individual. Or they have individual clarity that doesn't ladder up to a real vision or a real ambition. Because you can be really clear about — please take 10 steps to the left — but you're like, why am I doing that? Well, you know, just I thought it'd be nice to make you busy. Busyness becomes a thing. And so I think this is true with — it's almost even worse when there's no cost, right? With humans, at least there's a perception of — you and I both worked in intensive early career environments where, you know, telling the team to pull an all-nighter was not really a big deal. Right? Getting your AI agent to pull an all-nighter is like, well, what do I care? But these are real resources. And so I think one of the things people are going to have to think about is: clarity does not become less important when you have more resources — it actually becomes even more important. But I love your messy middle. I think you're spot on. Well, I want to talk about — what do you think is the leadership capability that will be totally obsolete in an AI world? Any thoughts on that? Steven Wolfe Pereira [0:35:30] I mean, where do we begin? That might be a whole other pod. Jason Baumgarten [0:35:34] Pick one, pick one. Steven Wolfe Pereira [0:35:35] I really do think that — again, not my words, just listen to any of the AI leaders. Pick your favorite — could be Dario over at Anthropic, it could be Demis over at Google DeepMind, maybe Mustafa over at Microsoft. Obviously Sam, you know — I think this whole idea of management is kind of cooked. The whole idea of this kind of command and control. And I know it's not every company, but it's a lot of S&P 500 companies — kind of carve out a lot of what's in the Bay startup culture — it's just fundamentally different. But the way large organizations that really have command and control structures, where the leadership decision is coming from the top — that just doesn't work in an AI world. And so I think what you're going to start to see — and we're already seeing it — like I keep trying to tell people: and again, I'm an AI optimist, but I'm also a realist. And so I feel like right now we're all, as a labor force, the frog that's in the pot that is boiling and it's getting hotter and hotter. And what you're going to start to see is slowly but surely more of these kind of mid-level to senior management jobs go away. They're going to get eliminated. They're expensive, they actually don't add that much value, and they actually aren't ready for the AI era. So if you don't understand how to use any of these tools, if you have no idea what an agent does, if you have no idea how to actually develop the workflows that agents are going to manage — what is your value? And I feel that you're starting to see it. I mean, literally yesterday, Goldman Sachs said that they are now going to start to do regular rolling layoffs. You saw Morgan Stanley. It's not just the tech companies — you know, Meta or Block or Atlassian — like you're starting to see it in all these different areas. It's starting first in tech and certainly in software engineering, just because that's the first thing to go fully agentic. You still need to have software engineers. But what you're going to start to see is the K economy really turn into the kind of K labor force, where everyone is either going to have to be a builder or a creator, and everyone else is going to be a consumer. And so you — as whatever position you are in the company — are you the builder or are you the consumer? Because if you don't know how to build and use these tools — again, it's an experiential technology — if you don't know how to use these tools, you might be seeing that pink slip sooner than you realize. Jason Baumgarten [0:38:24] I think everyone's going to be trying to find that place of how do I add value? And I think for a lot of leaders, going back to the superpower concept — like, what really is the thing that allows you to add value? What is that superpower? What is it that makes you really in your flow? And getting in touch with that is important, because then you can say, hey, does that superpower outlive an AI-first world, or do I need to repot it into a different environment? And you know, maybe we'll all be AI humanizers, where we just rewrite stuff that AI writes to make it sound less AI-like. But I do think that for many people in business and in leadership roles, they're going to have to think a lot about how they add value. And that's hard. Steven Wolfe Pereira [0:39:02] Let me give you a positive, you know, kind of take on what could be a future role. I think maybe a future role of a leader — someone in a management position — is actually managing the people that are managing other agents. Right. So what does that mean? How are you actually helping them understand how to either take the company's values and really embed them inside of agents? Right? How do you actually ensure that they're learning everything that the company truly stands for — the intent? You know, I think we all saw what Klarna did and how they got very giddy about AI agents for customer service, fired all their customer service folks. But then they had a huge backlash because AI agents, by definition, as you well know, are goal-oriented — do X and they won't stop until they achieve X. Well, if the goal was be efficient in customer service calls, they're going to answer and hang up and hang up. That's a bad customer experience. So understanding the intent is going to be very important. I feel like that could be a role that management starts to play. And again, back to governance — I think you're going to see a lot more of that, because how you govern all these agents, what agents have access to, what type of data, permissions, internally, externally — that is kind of a very rich territory. And we're going to need to see a lot of new roles be created for that. Jason Baumgarten [0:40:23] I think one of the things we're seeing is a lot of debate between what is an interim role and what is a forever role. And I think some of the things you're describing, there'll be periods of companies needing to adapt to AI and roles that are stood up to codify, to train, to adapt. And then the question will be, is that done, or is that an ongoing role? As the world continues to shift and as you have to continue to teach the humans and the agents what are your values — were your values the same as they were in 1950 or have they changed? So it'll be important to think about the timelines. I want to — I did a fun little exercise. I told you I was going to do it. We asked ChatGPT what is your leadership philosophy in one line. So I thought it'd be fun to ask you what you think of its summary. Your leadership philosophy is: the ability to drive fast, high-stakes innovation while building the governance structure and human judgment needed to do it responsibly. Steven Wolfe Pereira [0:41:18] I mean, that's not too far off — not too bad. I'm a little bit wordy for my taste. I think Claude would be a little bit more efficient with economy of words. But I do believe that governance is an accelerant. I think governance has historically been viewed as, oh, kind of in the risk camp. But as you know, we've spoken about this before. We at Alpha have kind of this two-by-two matrix that we really use to guide AI governance. On the X axis is internal and external and on the Y axis is risks and opportunities. And you need a four-quadrant approach. So if all you're talking about are the internal risks, you're missing the boat. You're not reimagining all the external opportunities that you have to reimagine your customer experience, the way you actually deal with partners, as well as external risks. And then, you know, the low-hanging fruit of course are efficiencies internally — that's an opportunity. But at the same time, you know, maybe all those savings you actually redeploy — because I don't know about you, every CFO I know, of course they would love to save a million dollars, but they love even more to actually generate $10 million in revenue. So how do you actually redeploy whatever savings to kind of get more growth? Jason Baumgarten [0:42:29] Yeah. And that's a holy grail. Okay, so then we asked it — how do you think? Right. Not the agent, but you. How do you think AI changes leadership? And it's — I thought this was quite eloquent. AI doesn't just give leaders new tools, it raises the standard for leadership itself — from managing people to governing systems. And I like that. Governing systems, because it really is what you were talking about. This is about what are the underlying systems that we're using to make sure that it's doing what we want and it's continuing our values and our judgment, not creating it. So what do you think of that one? Steven Wolfe Pereira [0:43:03] So I believe in being a systems-based thinker and I learned that from first principles — literally from the guy that wrote the book on parallel computing, Dr. Tom Leighton. Right. So he is the CEO of Akamai, the co-founder. But when they started Akamai back in '98, you know, he was running the lab for computer science at MIT. So so many of the AI leaders today actually learned from Tom. And you know, Tom is still going strong as the CEO of Akamai today. But what I learned from him is systems-based thinking. And so I think you need to have a systems-based approach to AI governance because it impacts every element of the enterprise. Again, I think the days of silos are done. You cannot operate that way. You have to think in terms of workflows and what becomes — just take your go-to-market, for example. You know, every single CEO that I hear on a public earnings call, they're being asked, what are you doing with AI? So they're all AI washing. Right. And they're giving examples of how they're reimagining the customer experience. Well, what is the pre-sales — if you will, the marketing? What is the point of sale — the sales or the conversion? And then the post-sale of the customer experience? Those are like maybe three different teams today. But with an AI agent, that literally is a unified agentic customer experience. So where does the marketing begin or end? The sales begin or end? And then the customer service begin or end? You can literally have an agent that does all three. So I think we need to radically rethink this. But that is a systems-based approach. Jason Baumgarten [0:44:38] I couldn't agree with you more and I think this really goes with your concept that organizational design is going to change radically. Even if we look at marketing — old acquisition marketing, retention marketing — it's crazy. We don't need that. That was designed because we had these limited constraints and we had to solve — Steven Wolfe Pereira [0:44:53] For the people. And go a step further. We have been marketing or selling or trying to acquire a human as a customer since the dawn of time. We now have a completely new customer in an AI agent. So when you market to a customer — if I'm trying to sell to Jason, I'm going to try to have really great storytelling. I'm going to have pretty pictures, my Instagram feed, all these things that will touch, move and inspire you. But if I'm trying to sell to an AI agent, I don't care about any of that stuff. I care about your data. So this is now a technical way to approach a new kind of customer. And agents are going to be the predominant customer that most enterprises will have in the future. Again, just wrap your head around that concept. We are entering the agent economy — and again I think that is so underestimated. It's so under-hyped. I don't think people realize how radically transformed the economy will be from agentic commerce. Jason Baumgarten [0:45:53] I used to have this joke I would tell when I was out to dinner with friends and we'd laugh about maybe overeating just a little and say, my dream in life is to have somebody else run around the block while I eat. Steven Wolfe Pereira [0:46:05] And I'm laughing. Jason Baumgarten [0:46:05] So I can't quite figure — Steven Wolfe Pereira [0:46:07] It out in food. Jason Baumgarten [0:46:07] But man, at work it certainly feels like we're headed in that direction — where we're going to be sending the agents off to do things and, you know, chilling while we wait for them to come back. But it is — the responsibility that comes with that is very real. And I think that issue of you still need the judgment — if you give garbage in, it's going to go buy stuff you don't want or don't need because you haven't been clear and precise as a leader in what you need. And by the way, I mean, we've seen this in software procurement over the years. You get a team to define what do we need to buy, and you get 78 pages of documentation, and then you buy something, you spend a year doing it, you put it live and somebody goes, I don't like it. And you're like, well, it's because it turns out you didn't boil it down to the essence of what you needed. Everyone just got a chance to add their wreck, and 78 pages later you're like, it should have been one or two pages. Fewer people might have been happy at the start of the process, but it would have solved the real problem we're trying to solve. At McKinsey we always said you spend a lot more time defining the problem you're trying to solve. Because that's usually where the thinking is the loosest. And then when you get really crisp on the problem, you're off to the races. Steven Wolfe Pereira [0:47:16] Well, again, it's inputs versus outputs. Right. I always tell folks, revenue is an output — that's not an input. Right. So when you think about it, the futures are already here, it's just not evenly distributed. And again, Ethan Mollick, you know, at Wharton, does such a good job of talking about the jagged frontier. You know, this is going to be all about diffusion. We're going to see it always as kind of patient zero is going to be all of the AI startups in the Bay — they're already agentic, they're already making the bets on who's the solopreneur to be a billion-dollar company with a ton of AI agents. That will happen. But how does this diffuse into the rest of the economy? What are going to be the enterprises that really start to adopt this? And then how do we really think of all of the challenges that the boardroom — both the directors and the management teams — are going to have to deal with: the issues of labor, who are you going to let go, how do you kind of bring in digital labor into the workforce, how do you think about narrative AI threats? You know, there's a really cool company called Blackbird AI and they're helping companies, they're helping countries — you know, I think NATO or OECD is like one of their new clients they just announced — like, how do you deal with narrative threats that literally start from AI agents and they do agent swarms and they kind of get a narrative about a company, about a director, about an executive, and that kind of spreads like wildfire. You end up seeing it pop up on CNBC, the stock's down 30%, and now you have an activist investor that is basically laying into this because all these things were kind of concocted. We haven't even begun to see the threats of cyber amplified by AI. So again, a lot of good, but also a lot of bad that you have to be aware of. Jason Baumgarten [0:48:58] Yeah. The dark side of what we're going to see is scary. I saw there was a clip floating around yesterday about a new anti-AI vaccine that you can go get at the hospital, and somebody just doctored a news clip and it looked like somebody rolling into a hospital. But the very official clip now read — instead of getting the COVID vaccine, it was the anti-AI vaccine. And this was making its way around the web. And I thought, this can be a scary few years. At the institutional level you may be able to defend, but I think at an individual level — and we see what's happening with our kids and social media and texting and everything else — it's really scary when you think about the implications for an individual, especially a child, a minor, somebody that is even — Steven Wolfe Pereira [0:49:39] More susceptible. But an individual — but now think about it back to the enterprise. How do we do security today? Jason, you're going to call your bank and how are they going to authenticate you? Your voice, maybe a biometric, some type of password. All those things are antiquated. You could literally replicate with just three seconds of listening to your audio — I could have a perfect replica and a digital twin of you. And so we have to radically rethink what privacy is going to look like in this world where everything can actually be deep-faked. And again, I think it's going to really come to a head when we see the midterm elections in the fall, because all that stuff is going to come. Jason Baumgarten [0:50:18] Well, the good news is we know what beepers look like. We're all just going to carry around our beeper and we're going to get the beeper code to — we're going to go old school with — Steven Wolfe Pereira [0:50:29] Right. We're going to go — Jason Baumgarten [0:50:31] LPs and bricking our phones. So. All right. We should do a speed round because they're always fun, of course. Five rapid-fire questions. Steven Wolfe Pereira [0:50:39] Oh, boy. Jason Baumgarten [0:50:41] And no overthinking. I'm going to cut you off. Steven Wolfe Pereira [0:50:44] All right. Jason Baumgarten [0:50:45] Best leadership advice you've ever gotten. Steven Wolfe Pereira [0:50:48] Oh, that's easy. From one of my mentors, Antonio Lucio. And it is: focus, focus, focus. Jason Baumgarten [0:50:55] Love it. The most important decision in your life. Steven Wolfe Pereira [0:50:59] That's another easy one. Marrying my wife. You talk about fit happens — having the right life partner makes everything else work. And so marrying my wife, Nuria Santa Maria, who is just an incredible technologist in her own right. But that was the most important decision. Jason Baumgarten [0:51:15] Yeah. It turns out fit happens is not just for work. Very true. Boardroom skill: most board directors overestimate in themselves. Steven Wolfe Pereira [0:51:24] I think it's their operating experience. Right. I think the whole idea of — and again, you live and breathe it — we're looking for a CEO, or we're looking for a CFO, we want someone that's actually had those type of experiences. I don't know if that's the only vector. I think they're underestimating the experience of builders — of people that have been living and breathing technology, either natively or they've learned it. But you need to actually balance it out with just the operating traditional skills. Jason Baumgarten [0:51:53] Yeah. So you're really getting at the context of the operating experience, not just having some operating experience. A book or idea that really, really changed you, honestly. Steven Wolfe Pereira [0:52:03] Right here. So Human and Machine by Paul Daugherty. He is the former CTO of Accenture. He's one of our Alpha fellows who is just brilliant. But he wrote this a few years ago and it was so prophetic in terms of where we're going. And so seeing human and machine in the workforce — I think that was incredibly enlightening. Jason Baumgarten [0:52:25] I am waiting for the sequel. I read that book — I think actually your recommendation. So fill in the blank: real career fit happens when — Steven Wolfe Pereira [0:52:35] When both sides want to make it work. I fundamentally believe that people say they want change — they really don't. I'm sure you have so many examples of this, Jason, where people want that innovator, they want that — but they really don't. They say it, but they don't really mean it. And so I think it really has to be that socket connection — where if it's a square peg, it needs to go into the square hole. Round peg into the round hole. It needs to be on both sides. It can't just be one person wanting the innovation and the other person pretending, feigning that they actually want it. Jason Baumgarten [0:53:10] That's so well said. We've spent a lot of time recently talking about derailers and third rails. And like, what is the thing that this person — that you know you need? Because they're bringing something you know you need for your context. But they're going to come with something that's going to make you a little anxious. It's going to be hard. It might even be their undoing. And is there something that is a third rail — that if they go over it, you're like, no, no, that is now not going to work. And one is about what they'll do and one is about the things in your organization or context that can do it. And I think often in these processes of courting — whether it's inside a company switching roles, or whether it's outside a company — people don't think about that. And I see it even in the boardroom where people say, I want that really disruptive thinker. And then they're like, could you kind of not talk, and, you know, you don't understand us. I'm like, no, I think they do — they're just asking difficult questions you don't really want to answer. And, you know, the reality is you don't have to change. You don't have to have a difficult thing. You can stick around, but you have to know your context. And fit is not static — it changes. Changes in life, changes in work. So you've got to keep renewing your vows, so to speak. Well, listen, it's always a pleasure. I love what you're doing with Alpha. For anyone who is not aware, please go check out what Steven has created. It's awesome. And thank you for being a guest. Steven Wolfe Pereira [0:54:28] Well, thank you for having me, Jason. And thanks for all the work that you do. Obviously, you work with so many different board directors, C-suite execs, and having someone that could go beyond a resume, go beyond just some superficial analysis, and really understand the human and try to get that fit — I just love everything you're doing with this pod. So thanks for having me. Jason Baumgarten [0:54:49] Absolutely.