Marketing School - Digital Marketing and Online Marketing Tips

What Eric Learned From Hosting A 9-10 Figure AI Operators Dinner

26 min
Jul 20, 20268 days ago
Listen to Episode
Summary

Eric shares key learnings from a dinner he hosted in New York for 9-10 figure AI operators, covering where companies are actually seeing margin expansion from AI and where they aren't. The episode also digs into a large-scale tech worker sentiment survey revealing burnout is surging and the top fear is being asked to do more for the same pay. Neil and Eric debate how compensation models need to evolve for AI-augmented workers, and reflect on entrepreneurship, parenting, and company culture.

Insights
  • AI-driven margin expansion is clearest in customer service (e.g. via tools like Fin) but largely absent in software engineering, where output gains are offset by higher salaries and continued headcount needs.
  • The dominant fear among tech workers is not job loss to AI but a 'do more for the same pay' squeeze, with burnout up 10 points in a single year according to a large-scale survey.
  • Founders report the highest optimism, job enjoyment, and AI excitement of any tech role, and the lowest burnout — reinforcing the 'never been a better time to build' narrative.
  • AI is enabling agencies and service businesses to deliver previously cost-prohibitive custom solutions (e.g. bespoke franchise reporting dashboards) for a fraction of historical cost, shifting value delivery rather than directly improving margins.
  • Compensation models for AI-augmented workers are unresolved; proposed frameworks include athlete-style supermax contracts, story-point-based pay, and upside-sharing on client revenue growth.
Trends
AI cost optimisation: companies over-indexing on expensive frontier models for simple tasks, creating unnecessary margin pressureCustomer service automation via AI is the clearest near-term ROI category for large enterprisesTech worker burnout is accelerating as AI raises output expectations without corresponding compensation increasesWorkforce bifurcation: AI is splitting tech workers into 'amplified' (50%) and 'destabilised or diminished' (19%+) campsCompensation model innovation is emerging — story points, athlete-style contracts, and revenue-upside sharing are being exploredAI enabling service businesses to deliver custom, high-value deliverables at dramatically lower cost, disrupting traditional project economicsTop engineering talent working longer hours (13-14 hour days, 6-7 days/week) despite AI assistance, raising sustainability concernsFounders remain the most satisfied and optimistic cohort in tech for the second consecutive yearSmart contracts and output tracking being discussed as future infrastructure for fair AI-era compensationToken cost management is becoming a critical operational concern as 'token maxing' by employees inflates AI spend
Topics
AI margin expansion in enterprise businessesCustomer service automation with AI toolsAI compensation models for high-output workersTech worker burnout and sentiment trendsAI workforce bifurcation — amplified vs destabilised workersStory-point-based developer compensationToken cost optimisation across AI model tiersCustom AI-built reporting interfaces for franchise businessesFounder wellbeing and optimism in the AI eraEntrepreneurship as a career path — risks and rewardsCompany culture, loyalty, and employee-founder relationshipsSmart contracts as compensation infrastructureAI productivity expectations vs sustainable working hoursRevenue upside sharing as an employee incentive modelAI operator dinner format as a peer learning mechanism
Companies
Goldman Sachs
CEO David Solomon was spotted at the restaurant where Eric hosted his AI operators dinner in New York.
Single Grain
Eric's marketing agency, cited as an example of using AI to build custom franchise reporting dashboards cheaply.
Intercom
Fin, Intercom's AI support product, was cited as an example of AI delivering clear margin expansion in customer service.
People
Eric Siu
Co-host sharing learnings from his New York AI operators dinner and reactions to tech worker sentiment survey.
Neil Patel
Co-host reacting to AI margin expansion discussion and sharing agency perspective on AI-driven service delivery.
David Solomon
Spotted at the New York restaurant where Eric hosted his AI operators dinner; noted for his side hobby as a DJ.
Lenny Rachitsky
His podcast was cited as the source of a large-scale tech worker sentiment survey on AI fears and burnout.
Quotes
"The number one fear in AI right now is not necessarily losing your job to AI — that's actually second from the bottom. The number one fear right now in tech is actually the do more for the same pay squeeze."
Eric Siu
"AI has created the opportunity for us to do more for our clients and do different things that would have been too costly based on their budgets."
Neil Patel
"Burnout surged 10 points in a single year, while optimism fell 6 points."
Eric Siu
"If my son had a choice from making millions and millions of dollars and being miserable or being a teacher and being happy, I'll tell him, be a teacher and be happy."
Neil Patel
"I think we need to reimagine compensation and I don't think it's going to happen overnight, but I think a lot more people are going to start to think about this."
Eric Siu
Full Transcript
2 Speakers
Speaker A

So, Neil, I'm in New York right now and I'm here for a couple of investor meetings. I am here. I'm speaking at two conferences. Spoke at Mozcon on Tuesday, which I've never spoken at before. And then today I'm speaking at a E Commerce conference, the social commerce summit that we spoke at in Manhattan Beach. So Stuart's doing another one. And then I also threw a dinner. Probably the highlight was the dinner that I threw on Tuesday and I wanted to share some learnings. And then I'm also going to prompt you with some of the questions I use because my Hermes came up with some questions that they're actually really good questions. And so this dinner was nine to ten figure AI operators. And then by the way, the restaurant I hosted at, I saw DJ Solomon. You know who that is?

0:00

Speaker B

No.

0:40

Speaker A

It's the freaking CEO of Goldman Sachs, man.

0:41

Speaker B

He's.

0:43

Speaker A

Remember, he's a dj. Oh yeah, yeah, yeah, yeah, yeah, yeah.

0:44

Speaker B

He gets made fun of sometimes. I'm not saying people should. I, you know, he should have hobbies. But I think Trump is one who makes fun of him for being a dj. It's, it's like DJ Solomon, I'm like, I don't know what music he's creating.

0:48

Speaker A

He was like, my friend and I were talking at the table and then I looked over, I was like, I was like, he's like, yeah. I was like, oh, it's DJ Solomon. And then like another investor came over. This, this place I hosted at it is so interesting because it's, it was like all investors and finance people there. Whereas in Miami where it's the same restaurant, it's not like that. It's very chill in Miami. So anyway, I hosted this dinner. You had people from, you know, we're talking about, you know, nine, ten figure companies like revenue wise. And I, I'm going to kind of just go down the list and share some takeaways and some questions. And then, Neil, we can kind of go back and forth here. And then I'm, I'm curious to see how your week went as well. Excuse me, I have a lingering cough as well. So first and foremost, Neil, one of the first questions I asked was where are you actually seeing margin expansion with AI? Which is a very important question.

1:00

Speaker B

Right.

1:52

Speaker A

Because we all like to talk about stuff we're doing right now. So do you want to answer that first? And I'll give you some answers and then I'll give you another good question after.

1:53

Speaker B

We're seeing margin expansion by AI, by cutting and you have to be careful which models you use because if you use the wrong models, they're very expensive. What we find is a lot of organizations use the most advanced models to do some of the most simplest things, which doesn't logically make sense with. They can use older models that are a fraction of the cost. But we see margin improvement from mainly cost cutting.

2:01

Speaker A

Okay, so the table struggled to answer this one. Okay, now one of the guys was like, was very clear for us, this is a multi billion dollar company. He's like, we're seeing margin expansion clearly in customer service. Right. So customer service, we're seeing that because obviously that one is, um, you know, you look at a fin, for example, it can handle a lot of these, these support tickets. And so you're actually saving, you're, you're, you're saving significantly there, but in other areas, not as much. Right. And so for coding you can say, yes, we're getting more output, but what ends up happening is you need to hire more engineers anyway. Um, so it's not like anybody has

2:25

Speaker B

created and paying more and pay them more. The good ones want more money. We're, we're not seeing margin expansion in developers.

3:00

Speaker A

Yeah. And by the way, like one, one topic I started to bring up last week was this whole thought around if all these people, let's say, Neil, you become more powerful with AI, but I don't, but you're able to do all these other things, how do we handle compensation with you? And this was a really, like I asked it again at this table, I was like, how are you guys thinking about compensation when it comes to the AI pilled people? And I might as well bring it front and center because I think a lot more people are going to start talking about this. And so one of the guys, he's like, yeah, I brought it up. And then the HR team's like, well, we already have these structures and all that and this is how it's going to be. And so he doesn't necessarily agree with it. He thinks that people should be compensated. AI pill, people should maybe be compensated like athletes. Like if you're LeBron James, for example, you get a supermax contract. Or if you're somebody else, like, I don't have a good name here, but if you're, I don't know, I was trying to give an Asian player name, but there aren't many of those either. So my point is, you know, when you look at an athlete, a superstar athlete, they get paid a lot more. So that's an interesting model. Someone else at the table, they compensate their developers just a lot of money. Right. And, and I'm like, how does that work exactly? Because their, their developers can make up to seven figures and they, they start at, at a good amount initially. And he's like, we, we charge them based on these, these story points. And I'm like, what's a story point? And he's like, a story point is like when you have a storyboard for a product, but we charge them per that. I'm like, isn't that just time and materials? And he's like, well, no. And he gave me some answer, but then I still interpret that as time and materials. So there's a couple answers here. But Neil, I'm curious to get your thoughts. We should react to this in terms of how we're thinking about it.

3:07

Speaker B

Yeah, just honestly, internally, we're not thinking about margin improvement from AI. We're trying to figure out, or not even trying to figure out where we're focusing our time and energy is how to get people more productive and do more. And I think the key is do more because when you're doing more for the customer within the same price point. Yes. You're using this technology, it helps you be more efficient. Maybe you don't need as many people in certain roles, but you are paying for AI and tokens, which can be costly depending on which model you're using. Net Net. We're not seeing too many margin gains and we're not seeing too much extra profitability because AI just being really transparent, we're able to do more that we've never been able to do before, but that costs more money. Right. So I'll give you an example we have. We work with a lot of franchises, and franchises are all different types. Some people are old people home. Some people could be like a dentistry franchise. Some people could be a restaurant franchise. Some people could be like a H Vac franchise. The list goes on and on. And what we found is this is obvious. Every franchise has its own needs. So we started building custom interfaces for franchises to make the reporting better, to make the experience better. Because some of these people have like over a thousand locations and it can be really painful to manage it all in traditional systems. So building something custom for each of them would cost hundreds of thousands of dollars, if not millions of dollars. And then you got to maintain it with AI. We're able to do this really quick, really cheap for a few thousand bucks or five, ten thousand dollars. Whatever it costs may be, including human time, customers are happier. It Makes life easier. Not necessarily meaning we're saving money, but life easier from a reporting angle, making us making less mistakes, franchises making less mistakes, people being more clear on what's happening, the results. And what we found is AIs created the opportunity for us to do more for our clients and do different things that would have been too costly based on their budgets. Of course we would have loved to do these things in the past. Client's not just going to fork over extra million dollars for a custom interface that's designed just for them. In most cases, from what we've experienced. And these kind of things haven't made us more money, they haven't really saved us any money. We're just now doing new things that were not financially feasible for our client, but now are. That's a big thing that we've noticed as a marketing agency because of the technology.

4:44

Speaker A

You know, it's interesting, I was listening to this podcast with Lenny's podcast, Lenny Rachinsky, I think is how you pronounce it. And basically they ran a large scale survey around tech worker sentiment, around AI. And so the number one fear in AI right now is not necessarily losing your job to AI, that's actually second from the bottom. But the number one fear right now in tech is actually do more for

7:32

Speaker B

the same pay squeeze.

7:59

Speaker A

So most workers are. Because you just mentioned do more. Right. And with these people that are very excited about. I think you have 41% of people that are net excited about this stuff. Their whole thing is burnout more. Right. And what's interesting is that the founders that were surveyed, they're having the most fun ever, but they would never recommend founder as a job. And that's funny. That's not really funny. But you and I, I don't think we'd recommend founder as a job to the vast majority of people because it's really hard. But we recommend it to maybe like 1% of people. Right, But I thought that's really interesting that you brought up the do more thing. And it turns out that the number one fear in tech right now is the do more for the same pay squeeze.

8:00

Speaker B

Yeah, dude, you don't have kids yet, but would you ever want your kids to be an entrepreneur? Honestly, I'm just curious.

8:38

Speaker A

Nope. Oh, actually, hold on, hold on, hold on. Maybe my son, but I wouldn't want my daughter to go through that, which is, I feel like that's how you've set up yours. Can I. Is that kind of how you feel?

8:45

Speaker B

So both my kids are great. My daughter, she's young. She has made up her mind that she wants to. She's really well educated for the age of seven and so is my son at the age of five. I'm not saying that means they're more brilliant or anything like that. Just being quite frank. When you have more money, you can pay for private schools and tutors and all these things that help, that help your kids get ahead in life. Right. And I wish everyone else had those opportunities as well. But I'm just speaking real here. My daughter right now at the age of seven, a lot of times she tells me, like, what do you want to do? And she has some ideas on what she wants to do. But then she says, in the long run, I want to have a family. I want to raise them and I want to be home cooking for them, taking them to school, you know, and helping them be the best kids they can be and be successful when they grow up. Right. And some people are going to hate me for, you know, for what I just said there. But I'll also say one other thing. I believe being a mother is the hardest job in the world. I think being a stay at home mom is much harder than working 9 to 5. This is just my personal experience. Even as an entrepreneur, I think what I do is much easier than raising kids. I cannot do it. I think it's way easier to get on a plane, speak at a conference, do some meetings, go back, you know, I get to watch Netflix. You're a mother. A lot of these mothers have it even worse. They have to work 9 to 5 and they have to take care of the kids at home. I get. They may get help from their significant others, but check this out. If you're a mother and you're sick and then your kids are sick, they don't care. They still expect you to help them out. Like, you never have a day off in your life. So my daughter said that if that's what she wants to do, more power to her. And then she asked me, I was like, so what are you gonna. Or technically asked her, I said, what are you gonna do for money? She's like, I'll work for a little bit. But then, you know, I'm gonna live with you. I'm like, oh, you'll live with me? I'm like, I would be very ecstatic and happy if she listens to me. I'm like, you sure you're gonna live with me? You pinky promise? I'm like, don't change your mind. And literally, before she pinky promised, she's just like, well, actually, it may be easier if I live next door to you. And then I said, how are you going to afford the house? And then she says, you love me, right? And I'm like, yeah. She's like, you'll help me. And then I was just cracking up for my son real quick.

8:59

Speaker A

If you want to acquire customers faster and more efficiently this year with the latest strategies and tactics, then check out singlegrain.com, that is my ad agency again, www.singlegrain.com. check it out and if it seems like a fit, we'll get in touch and help you with a free marketing plan.

11:40

Speaker B

If he wants to be an entrepreneur, I would be okay with that. He's really into numbers. Like, for Christmas he asked for things like a calculator. I wonder where he got that from. Yeah. So if he wants to be an entrepreneur, I would be really happy. I wanted to name him Neil Patel the second, you know, and that did not happen. It wasn't that I couldn't name Neil Patel the second. I wasn't sure. I was like, should I? Should I not? And then I settled on. I'm like, ah, let's just. I think it's better not to.

11:56

Speaker A

Because imagine if Michael Jordan named his kid Michael Jordan. I think he did, actually. That's. That he had. If he doesn't live up to the name, it's tough.

12:29

Speaker B

Correct. And I'm not saying I'm an Elon Musk or I know I'll never be that successful, but still, I don't want my kids to believe that they had to do what I did because. And I don't think it's the best quality of life. What I think is actually better, like, you know, my son's like, maybe I'll get to be a scientist or I'll get to work with animals. Because he loves animals and animal facts and he's young and I don't know what he wants to do. But I do want them both to work and I do want them to understand how hard it is. Whatever they decide in the future is up to them. But if my son had a choice from making millions and millions of dollars, dollars and being miserable or being a teacher and being happy, I'll tell him, be a teacher and be happy and go live in, like, Missouri or something like that. Now, if he wants to be a teacher in Los Angeles county, where he probably wouldn't be able to afford a home because it's just sad but true. And, you know, he had kids and he would be Next to me, I would help him out, because what's the point of me working all this hard all these years to not have my family next to me?

12:37

Speaker A

I will come back to the compensation thing in a moment, but my take on it is, the way I see it, is I would want my son to go through it because, you know, it's. I see entrepreneurship as like a rite

13:44

Speaker B

of passage type of thing.

13:56

Speaker A

And you learned a lot. I think it's a great learning machine. I don't think I'd want to put my daughter through it. That's just my philosophy. And I do agree that being a mom is the hardest job. Um, I looked at my mom. She had to be the breadwinner, and she had to be the mom. That's ultimate double hard. Right? I think your mom was like that, too.

13:57

Speaker B

Yeah, same.

14:13

Speaker A

And so I. I wouldn't want my daughter to have to do both. At the end of the day, I think one's already hard enough. Um, so that's just us. But going back to this survey for a minute, Neil. So I'm going to read some of these things, and then let's talk about AI and compensation. Okay. So you have the. The number one fee right now is doing more for the same pay. Okay. Most people in tech would not recommend their own role to someone else entering the industry today. Okay. Designers and researchers scored the worst. Founders came closest to neutral. That means you and me are closest to neutral. Everyone else is there not recommending their job to someone else. That's fascinating to me. Now, the one final thing I'll call out here. Let's talk about the compensation piece is AI is splitting the tech workforce in half. So when asked how AI has shifted their professional identity, 50% of respondents said they feel amplified. Okay, I feel like that. I think you feel like that, too. So we feel more capable, more productive, more excited about their future. Right. The other half feel like the role is being redefined. So 27% that they're feeling destabilized, 14% or that they've been diminished, 5%. So I would say maybe 19% of that is fairly negative. I think redefined is not necessarily negative. And which category you're in correlates with your career. Optimism, burnout and layoff worry. And then the final thing final, final thing for real now, burnout surged 10 points in a single year, while optimism fell 6 points. I think burnout the year before was like, maybe like 44%. Now it's like 54% or something like that. So I. We've Said this on the podcast before that. My number one concern, especially with the AI pill people is burnout.

14:14

Speaker B

Yeah, I, I, I definitely can end up seeing it. Um, and I think things will change within a few years because I don't see people going this hard, especially in certain roles like engineering where they're expected to do a lot and work a lot of hours and crank out more. I know the top players are getting paid well, but dude, you know, they're saying, oh, the engineers don't have to code as much. A lot of the A player engineers that I know now are working six to seven days a week and they're working like 13, 14 hour days. It's just not sustainable. Just being realistic. I don't think it's sustainable, especially if you have a family. And I just think if you're not getting enough sleep, you're not as productive during the days that you're working. I don't care if you're using air or not, you're just not as productive if you don't get enough sleep. But I think things will change. It's just a question of when.

15:43

Speaker A

So it says also founders are still the happiest people in tech for the second consecutive years. Founders score highest on optimism, job enjoyment, AI excitement, and lowest on burnout and layoff worry. There's never been a better time to build narrative holds emotionally, at least for now. So anyway, going back to compensation, Neil, so we've talked about, some people look at it as, oh, maybe you have to pay your superstar athletes. Okay, the second piece is, okay, maybe you're awarding people based on these story points or maybe you're giving people a percentage of upside based, let's say clients paying 10 grand a month and then they upsell to 100 grand a month. Then on that 90 grand difference, maybe you're getting like 10% of that. Right. And so I think we need to reimagine compensation and I don't think it's going to happen overnight, but I think a lot more people are going to start to think about this and when we have better tracking when it comes to outputs or outcomes that people are delivering and we're able to kind of tie things in with like smart contracts and things like that, then I think we'll, we'll people will be more satisfied with their compensation. Because I don't think, Neil, you're necessarily wrong for saying do more, that that's becoming more an expectation. But I do think compensation needs to adjust as well. Otherwise it's the rich are going to get richer and then you know the pitchforks are going to come out.

16:34

Speaker B

When you say the rich are getting richer, I agree with you. It has to be fair for both sides. And I'll give you example of this in a bit. But when I say that if you're doing more, but your token costs are through the roof and it's very expensive for the business and their margins are going down, that's also not fair to the business. Assuming you're doing what's required as one individual. One individual. There has to be a balance. And what I'm getting at is it's like if you're using AI and you can't replace any other people and the customer's paying the same amount of money, something's going to end up breaking. And I think companies are realizing that they're spending too much. A lot of these employees are token maxing, which I think is over now or is going to be over soon enough. And it's all about just making it fair on all sides. A real life example of this is my buddy works for a software company, and the software company focuses in the health service space. So him and their team members are on a thread. They're really friendly and they were complaining how they barely got enough of a raise to even meet inflation. So they're like, and this was just a, a raise for standard of living. This wasn't a raise for promotions or doing better. This is just a standard of living increase. And they were all complaining that they barely got enough money. And the first thing I asked was, does this, does your company that you work for have investors? They said, no. I said, how was the financial performance of the business? And they're like, oh, we've been struggling over the last two years. We've been doing worse. So I look at that as, so the founders are helping you out. And I asked them, I'm like, do the founders live lavish lifestyles or anything? They're like, no, they're putting all their savings back into the company to keep it afloat. And I'm like, so you guys are pissed off that they barely gave you enough money to meet the inflation rate, but the founders are putting their personal livelihood into the business so that way everyone gets paid and the founders aren't taking any salary and they're not living lavish lifestyles. And I'm like, you also have to look at it from their point of view and they need to look at it from your point of view. But they're just trying to keep everyone, you know, doing okay with the job. And I said, hey, have they thought about raising money? They're like, yeah, they tried. They weren't able to do it. So they're just trying to keep the company afloat and not fire anyone. And I'm like, look, that's pretty admirable, right? Not a lot of founders were willing to do that. They would just fire and just think for themselves and do what's best for them.

17:46

Speaker A

Neil, let me take the other side of that, just to experience, share. And then I think you'll find this. I think you'll understand where this is coming from. So, you know, whether I've put millions of dollars back into the company a long time ago, right. Or whether there's a period of time where I didn't get paid. Nobody gave a damn if it was my personal sacrifices or if I didn't get paid a dime. But they do give a damn if you are living a lavish lifestyle, right? Which, which, like, it's. It's funny, they don't give a damn when, when, if you're sacrificing. And I won't even put this on anybody. I think it's just human nature, right? Human people tend to look at their own situation, and then when it's like, oh, if the founder's sacrificing even more, it's like, well, my situation still sucks, right? So I think I understand where you're coming from, but I just kind of sharing my lived experience here. The vast majority of people won't care that you've done that, dude.

20:25

Speaker B

I totally agree. So, you know, and I'm not saying that people should care or shouldn't care or shouldn't switch and try to get a job. Uh, it's just finding not only the right balance, but this is why I think it's really important to hire people who are a really good cultural fit. If you don't, I think you just run into issues where they don't look at the whole picture or care as much for the company, and the company doesn't care as much for the people, it needs to be both. Like, I think of company as a family. If the founders don't care for the team members and the team members don't care for care for the company, it's not going to work out. And I'll give you a real life example of this. I had someone whose kid was sick, like, really, really sick, really badly, and I'm happy everything ended up working out. We told him, don't work, just spend the time with your child. Right? No parent should ever have to see their child be really sick or even potentially pass away before them. It should be the kids seeing the parents pass away and not the other way around. And we didn't care how much time they took off or whatever they needed still pay them. Right? But you should support your team members when they are struggling and when they're really in a time of need. I've also seen people abuse this and say they're sick. I think you went through that experience where someone said they were sick and they took another job and they were just milking money from you and trying to take your clients and create their own business that competed with you on the on the side that did the same exact thing while they're getting paid from you and they were just blatantly lying right? To me that's just unethical and people shouldn't do that but yeah, you can see all sides of it. That's it. Thank you guys for listening in. We'll see you tomorrow.

21:24