Is AI Coming for Your Job? Maximos Lih Says It's Not

John Golden: Foreign. Welcome to another Expert Inside Interview. My name is John golden from Sales Pop Online, Sales Magazine and Pipeline or CRM. Joining you as usual from a sunny San Diego. And today I'm delighted to be joined by Maximus Lee. Was up in uh, California, is up in Oakland, just up there, up the road in California. How you doing Maximus?

Maximos Lee: How are you?

John Golden: Good. You spent nearly a decade as an operating partner at Google Ventures, working across the portfolio with companies including Uber Slack, Blue Bottle Coffee and Flatiron Health. You're the founder and CEO now of Emboldened llc where you coach executives and founders on building leadership systems at the intersection of people and revenue. And today you share a framework that challenges how most organizations think about AI. Um, so after nearly a decade at ah, Google Ventures, I mean you worked with a lot of companies at different stages. What did you learn about what really drives outsized results when new technologies enter the picture?

Maximos Lee: I think that what is interesting about looking at ah, my time now in the last three years versus what it was like in the GB portfolio isn't the technology. All of the companies that came into the portfolio, the ones that I work with today, start with technology as the differentiator. Right? These are incredible founders who have left great, great careers at ah, Google, IBM, Oracle to go and start companies because they had a technology hook. Um, what I think is different about where it is today is where does AI fit into that? How does AI actually change everything that we know about go to market? And um, I have this very interesting statistic that I learned about a month ago, which is that ah, uh, did you realize John, even as we speak Today, more than 51% of online traffic on the Internet is now agentic, which means the Internet is now agentic. Right. So how does that actually impact the way that we think about any kind of growth? And uh, whereas I would have said in 2023 when the MIT report was coming out, the sales leaders were saying like, oh, is AI going to cause layoffs or replace my people? Not really the question that they're asking these days. The question is what, what am I going to do if I spend all the money by all the tools and deploy everything that I'm supposed to, But I literally can't tell if my teams are better or more productive. Yeah, right. That's actually. Right. And that's where the problem becomes very, very different and much more interesting. But it's also the same. It's still how do you develop leadership to do true. Go to market?

John Golden: No, absolutely, absolutely. And you saying like, um, a lot of companies are seen to be adopting or uh, implementing AI in the same way. What is the common way everybody's doing it and why should they be doing it differently?

Maximos Lee: Yeah, I would say that, you know, in my coaching work, typically three types of leaders come in where they're like, I don't know if I'm getting the roi, you know, tell me what's, what we're doing wrong. Right. The first model is of course the outsourcer. These are people who just. Let's make it a math problem. Right. If my AI customer service agent is taking on 50% of the calls, I lay off 50% of the people. That's totally very just. Right. The um, problem is that they forget humans are IP sometimes. And in sales especially like the relationships are an ip. Sometimes buying behavior is emotional, sometimes you miss out on the moat when you're just using the AI into. The second model is people who are thinking about it in terms of like augmentation.

John Golden: Mhm.

Maximos Lee: Call them the delegators. Hey, let's just give everybody a 500 chat GPT budget, go forth and be more productive 10x your own job. Right. And the problem is if you map your AI strategy to activity and not to judgment, you've literally scaled all of your.

John Golden: Right.

Maximos Lee: So I sort of like um, make this comparison. I used to have an analyst who wrote five beautiful reports every month. I tell them to write 50 reports every month. I, I don't have time to read 50 reports or to find 50 important problems. But they have like used their ChatGPT budget to create things that are actually waste on the other side.

John Golden: Yeah.

Maximos Lee: And then of course the last one, the one that I have the most empathy with other people who are just feeling very anxious. I'm not trying to be dehumanizing, I'm not anxiety. We're going to try and wait and see. We're happy to follow, you know. But now the agents are everywhere. Right. But now the Internet is agented. Your landscape has not waited for you to be in committee or to do the research. And every day is an accumulation of harm and an accumulation of your competitors advantage over yours. Now,

John Golden: uh, because I do agree with you and sometimes now people are getting so kind of wrapped up in AI that they're as you said, like producing loads and loads of insights and research, constant flows of it and then tweaking and then going back and it's almost like there's a danger there that you end up in the diminishing returns category and you're over there investing a ton of time for instead of actually taking maybe what you've discovered, stress testing it, going off, sanity checking it, some other people then decide, and then putting a stake in the ground. Because if you just continue going and iterating, you're never going to get out of that.

Maximos Lee: Right. Well, in fact I was, it occurred to me the other day when I was trying to explain this to a, uh, company that we've actually seen a lot of this model before when the automobile was first invented. And when I say automobile, everybody will think, oh, you're going to tell the story about faster horses. It's actually not the story that I'm going to tell. The other thing that happened in the early days of the automobile was the car companies would send salespeople out to farms, right? And they would say, hey, if you buy these cars, you're not going to need horses anymore. Your life is going to be totally different. And the farmers would ask, how many horses are we talking about?

John Golden: Mhm.

Maximos Lee: Which is how we came up with the language of horsepower. So that these salespeople can go back and say, like, you had 100 horses, I got 100 horsepower car. It's like a swap. But the problem, and this is the problem that we're seeing right now, is you, if you used to have like a carrot farm and you had 10 horses to 10 wagons that would go to 10 different markets and that was 100 horses and you bought one single hundred horsepower car.

John Golden: Mhm.

Maximos Lee: You've actually not done any replacement because you can't, you have not actually translated the language of procurement, the language of utilization. They don't naturally flow into one another. For you to actually know how to use the car properly, you have to actually start thinking in terms of routes and not in terms of wagons. Whereas you used to just be able to think in terms of wagons. And there are people who are like, now I have rotted carrots because where I ran out of gas because there was no gas station. Right. Or you have people who overpaid. I had 10 wagons, I bought 10 hundred horsepower cars. They go to different places. I haven't produced any more carrots. Maybe I got to market faster, but did I really gain any? These things are not, they're not directly translatable. And I feel like, oh, that's exactly what we've done with AI. We've confused the language of, uh, procurement with the language of utilization. And that's not a sales problem, that's a leadership problem. You have to stop thinking in terms of wagons and you have to teach People to think in terms of roots. And the really, really smart people start thinking, well, I'm a small carrot. I don't really need any cars. What if I built a gas station on my farm, right? What if my horses went out to people who couldn't go to my markets before? And I, I got out of the carrot business and got into the gas station business. Now all of a sudden, there's some real value. And those people who are thinking in terms of roots are going to benefit, right? In a completely different way than people who just spent the money buying 10 cars to replace 10 wagons.

John Golden: No, absolutely, absolutely. It makes total sense. And you've, you talked about Ikea. They had like a 53 AI failure rate, but they turned it into a 1.4 billion revenue opportunity. How did they do that?

Maximos Lee: Yeah, yeah. So ikea, and this is, it's such a beautiful story of, uh, once again, right, like leaderships. That's not thinking like the outsourcer, the delegated or, or the anxious person. But Ikea, always on top of trends. They created an AI chatbot for customer service. It was called Billy, named after the, uh, Billy bookcase. Within six months of launch, it had started to take over 47% of customer service calls. Those very routine, where is this in stock? Where's my shipment? I want to order this thing. Um, 8500 call service, uh, agents were varying actually, in a traditional model, in danger of being laid off. But they had a VP of data and a VP of learning who kind of said, well, wait a minute, if that's 47%, what's going on with the 53% M. Right. And this is actually where I'm like, this is where wagons and routes make such a difference. Because the people who think in terms of wagons would have asked a very different question, right? They're asking questions that, like, what Meta is asking 53% failure. What do we need to do to train the model to be more perfect? Right? Like, you immediately go into model reliability as your first answer. You don't start. Start to think like, where is the failure of the AI an actual business opportunity? But IKEA asked that question, where is the failure of the AI business opportunity? And they discovered, you know, without needing to do a lot more work, that those were customers who wanted interior design advice, Right? They wanted people who could understand, like, if I've got four kids, like, how do you make my home comfortable? How do you make my home warm? What IKEA hack website should I be looking at to repurpose these things? And so uh, instead of laying off anybody, they repurpose, right. They reskill those 85 customer service agents who were already pretty good at talking to customers who obviously understood inventory. Right. Who like to listen and help, and they created a remote interior design business. Within the first year, $1.4 billion represented 3% of global revenue, estimated to grow to 10% by the year 2027. Yeah, right.

John Golden: Yeah, that's, that's an incredible, that's an incredible story because, because a lot of, A lot of people, um, right. Know, a lot of leadership are, are approaching this as, how do I save money first?

Maximos Lee: Yes.

John Golden: You know, uh, and, and, and that is leading to, you know, some strange decisions, some reversal of decisions, some m. Uh, unexpected results.

Maximos Lee: Yes. Efficiency is a commodity. You start to make headcount, right? Like tooling decisions based off of efficiency with the same tools that everybody is going to be paying 10 or like 10 bucks more or less for with the next, you know, three years. You have built your IP on something that is not, that will not bear weight. Um, and so I love that instinct of saying, like, oh, what is the failure teaching us about the business opportunity? And it is exactly right. Like the people who decided to get out of the carrot business and to build gas stations will build a completely different kind of moat. And it's not like, oh, I like 10% activity from efficiency. Right. It's like, oh, I got into something that's a completely different line of possibility. Mhm.

John Golden: And so, I mean, like you said, I mean, it really requires leaders who look and say, okay, here's what the machines can do.

Maximos Lee: Right? Uh-huh.

John Golden: What do we need to build around it?

Maximos Lee: Right? Yeah. I mean, and I think that what is really important is that, you know, you know, I know we have a lot of, you know, owners and practitioners and leaders and executives that listen to this podcast, right. Like, we, we are not equipping our sales leaders to understand how much more complex their role is. Anytime that you're doing strategic workforce planning or headcount planning, or even performance and efficiency work. And that's not just. I think you would know this, right? All the way up until 10 years ago, if I was working at a company that was like growing B2B SaaS and doing demand gen, they said, interview this VP of sales, interview this Director of Marketing. All I really cared about was past experience, executive presence, right? Like, and like, business acumen. Those are the three most important things today. You got to do a little psychology, you got, uh, to do a little therapy, you got, uh, you got to do policy. Right. And you got to think about AI and you got to think about, you know, future proofing your business and, and you have to build in resilience and you have to care for yourself. So if there's like the expansion of all of that, it is something that we've not built equipping or coaching, you know, to be able to address. But roles are expanding in all these really complex ways with very little calibration on. Like what exactly is the compensation structure of the future? Right. If everybody that I hire is going to own four agents, do those four agents belong to my company or do I belong to that person? And then do they get to command three headcounts like cost? Because you're saving me three people, junior people that I didn't have to hire. Right. Like all of this expansion that's happening with regards to how we actually think about, you know like wagons and roots is not being talked about enough. And the easy answer is to go that outsource. Right. Let's just make it a one for one swap equation because at least there's some math that I can stand at the end of it. But, but if that's the thing that's actually going to destroy your business. Yeah.

John Golden: And because we always seem to, we, we love to jump to implementation, don't we? Like and skip over the, the actual doing the proper, maybe strategic research and seeing what are the, what are the likely outcomes or where are the likely points of failure. Now we're so caugh speed and urgency that we jump straight to implementation.

Maximos Lee: Right, Right. And I think, you know, you mentioned systems design. I'm a big believer in systems design, especially for something like the agentic age. Because we just start to think, oh yeah, if I have everybody in my company training their own agents, they're going to talk to each other and they're going to mess each other's stuff up. Right. And when we try to move fast and we start to deploy agents before we really know what we're doing or what we, what, what quality control looks like for those agents, we've actually created more vulnerable for spam and fraud and like, and that's not one of those things that's uh, going to get better with waiting. Um, you know, and it's not going to be protected with more or less experimentation. Right. It really is starting with like, okay, so, so let me just assume that the AI does not have judgment and that the top 20% of my people have the judgment that I actually want to use to train. Like do you Know who those people are and do you know how to actually motivate them? Right. To be able to curate and build the things that are actually going to say I'm getting better at agents but not at the cost of more fraud.

John Golden: Right. Yeah. And, and also there's a danger, isn't there too though, that, that leaders start to communicate to their, their organization, their employees, like, you know, yeah, you're doing okay, but like we're, we're watching because as soon as the AI tools start to do some of your work, we're going to replace you.

Maximos Lee: Yes. Oh my gosh. The amount of anxiety around this space really feels ridiculous. Um, and I, once again, right. There's so little support infrastructure. There's so little support. There's a lot of demands to do more with less with no understanding of like what is that even supposed to mean for the business. So I do feel like that's one of those places where, um. Right. Like whether you are in the right business that you're trying to get your AI to do becomes the difference between, hey, this is a new business line, that's $2 billion. Yeah. Versus 8, $500 in layoffs and then like a bad defamation article.

John Golden: Yeah. And for me that's the real, the real danger is that short term thinking and rather than the strategic think, you know, thinking. And, and as you said, I mean, do you want a company later on that is that really at the end of the day is run by a lot of third party tools.

Maximos Lee: Yes. Oh my gosh. Yeah. And the more international companies that I'm working, uh, working with, where we just have exposure to, the more they're concerned about AI data sovereignty and over dependence on the United States infrastructure. So that becomes more and more, uh, complexity around that. A very practical thing that sometimes I will ask my clients is when you started using the tools which cost X number of dollars. I'm not even going to ask the multiplier whether you got anything back for that money. I'm just going to ask what can you do now that was impossible six months ago. And if the answer is no, but we're moving faster, then I guarantee you, you have scaled your bottlenecks. Right. As much as you scale your activity. Because the whole point isn't that AI is going to make uh, a faster human. Right. The AI is supposed to be, you're right. Like in maybe Elon Musk or you know, venture terms is going to make a more evolved human. Right, Right. Uh, so, so that, that's a really different kind of thing. Where I thought, oh, is there anything that you can do now that was impossible? Like, well, I, I didn't have to. I hired two fewer people. Right. Like, like, oh, well that's not.

John Golden: Yeah. And, and it's like, and what did that, what did that give you? You know, what was the outcome? What was the benefit of that? Like, is it just a cost saving or did, was it actually benefiting, um, your, your, your customer or prospect in any way?

Maximos Lee: 100%. Yeah, was like, oh, I didn't hire as many people, but now I understand my customers less because any feedback that's contained within these AI engines is not easily minable.

John Golden: Right, Right.

Maximos Lee: I actually have gained less competitive advantage or intelligence by this over reliance and blind faith that this tool is working on it, the things that it's supposed to be doing.

John Golden: No, absolutely, absolutely. Um, and, and tell me, what about those, what about those people who are on the other side of the fence, right, maybe who are fearful of implementing any AI and they're just sort of paralyzed right now.

Maximos Lee: Right. I think that those people, I mean, I think that they probably still are paralyzed. Right. Um, the good thing is that they're not making the mistake of putting a lot of money into efficiency that's just going to be commoditized. But the bad thing is they are, uh, the first people to fall to fraud. Right. First people who get scammed and they actually make them businesses so much more vulnerable to an inability to understand how everything has shifted underneath their feet. And I think that, that once again, right. So if you just keep asking the question, what is my business doing? That was impossible six months ago. Right, Right. And like the, if you don't do anything with AI or you don't do anything with technology, right. It's like, well, it's impossible to keep up. Right. Well, at some point that will become your answer. Like it's impossible to keep up. Um, and the people who are just like doing democratized 500 chat GPT but just everywhere, like what was impossible is that one, it's impossible to keep track. And neither of those things actually generate any running real or future proofing.

John Golden: Yeah. So what would be your, what would your final piece of advice to, to, to leaders on, on how to make sure that they can win in the AI era. And especially those who maybe feel like they've stumbled a bit or maybe they're a little disillusioned or again, maybe they're just, you know, a little bit, uh, anxious about it.

Maximos Lee: All right. Yeah. I find that when I talk with, uh, you Know my clients, when I tell them it's the failure of the AI that sometimes provides the most opportunity that gets both sides very excited. The people who have been sort of like a little bit suspicious are like, it's about the failures, right? Like, like, and the failures bring you back to a people and they, they get very excited about that. It feels much, much less threatening than needing to play this AI game. It's actually trying not to be tricked by the AI. And then the people who are really thinking about this in terms of like a one on one swap kind of, uh, perspective, it also allows them to actually feel good about being leaders. Right. Um, because if you're a leader and all you're going to do is basically own a bunch of machines.

John Golden: Right? Right.

Maximos Lee: Like, what's the point of being an employee at a company? I might as well start a consultancy. Right? Like, there's all of those things about the employer employee relationship where I'm like, yeah, maybe AI is not coming for your job. Maybe the AI is coming for the reason to stay in a job. And once again, when we ask the question, like, what is the AI failure teaching you about your people? Uh, and how is that helping you identify your best people and new opportunities that can be leveraged by them? Uh, that's, that's a leadership mental framework that people really can get. At least good leaders. Right? Good leaders want to empower their best people. They get really excited about that and that becomes a much, uh, it becomes a much more exciting topic for us to explore and it always pays off in dividends.

John Golden: Yeah, fantastic. Well, listen, Maximus, thank you so much for all of these fantastic insights. All of Maximo's information will be below this video. But before we go, please do remind people about what you do.

Maximos Lee: Yes. Uh, I would say, you know, I work with leaders in small teams, less than 50 people. I call that series A. You are really the voice of your culture. And what we need to do is to teach everybody at your company what good looks like. You hold a standard for that. Um, and in those cases, you don't really need like a full HR person. You shouldn't be hiring to the point where there's a lot of hash, HR complexity. But you know, I go and act as a head of people for that. And then for anybody who is growing their team from 50 to 500, how do I actually democratize culture, standards of good and start to empower my next generation of leaders, especially as their roles are expanding into these unknown places? And for those kinds of companies, I do a lot of coaching work. So, uh, if you're one of those people, this is exactly the kind of clients that I'd love to help program.

John Golden: All right. Fantastic. Maximos. Thank you so much. Thank you for watching, listening, and I will see you all again very soon. Thank you.