NVIDIA's Jenson Huang's unorthodox leadership style, and How AI agents will change marketing

Patrick Collison, the CEO of Stripe, did an interview with Jensen Huang from NVIDIA. And I'm going to read some takeaways from this interview that they did. You guys, I think you can watch the Stripe sessions on YouTube if you want. I don't know if the NVIDIA one is available to public yet. But number one, you probably know this one. Jensen has 60 direct reports. 60, 6-0 direct reports.

That's a lot.

Yeah. And so, you know his logic for that? His logic is all my people are very capable and they're all capable of leading something and just basically making something happen from start to finish. And so, I don't need another layer. I just go to them directly and we just get things done a lot faster that way. 2, he doesn't do one-on-ones. Interestingly enough, on Monday, I did a podcast interview with the CEO of Eight Sleep, Matteo. He doesn't do one-on-ones either. You don't do one-on-ones.

What do you mean by one-on-ones?

Oh, see, Neil doesn't even know what a one-on-one is.

No, I do. But then, it can be construed as very different.

A one-on-one in In a typical company, a one-on-one is like Neil and I meeting. And let's say I work for Neil, for example, right? So, how we do one-on-ones is I would bring my agenda to you. I'd bring probably a set of problems and issues you want to talk through.

No, no, no. I get the normal one-on-one. What I'm more so asking is, let's say if there's a problem with the chip and he's dealing with someone who's in charge of chips, does he actually go meet them one-on-one and discuss that?

Or is that— Yeah. So, he doesn't have a standing one-on-one.

Got it.

Yeah. So he'll say, do a one-on-one.

No, no, no, no. I thought you— I thought you meant he never meets with the individual one-on-one.

No. And then Mateo, same thing, right? So he doesn't have any standing one-on-ones anymore. He just has 2 hours free every day for office hours. And then that's when he just calls people or people call him. And if you want to do a one-on-one with him, this is interesting. If you want to bring an idea to him, you have to write a memo out, which is kind of similar to what we do at Single Grain. If you want to bring an idea, you're forced to write it out and it stops a lot of people in their tracks from bringing half-baked ideas.

The Amazon philosophy.

Yes. I'm going to continue on here. So number 3, All feedback— this is interesting. I think you would love this. All feedback is done in group settings. So, there's no like, you know how Warren Buffett is like, you know, criticize privately, praise in public type of thing?

Yeah.

He's just like, I don't— like, we have trust and we just get it out there. And that's that. So, and I feel like that's more convenient, but, you know, to each his own. Number 4, he doesn't fire people. I don't have enough context here. That's interesting to me. I can't imagine a company where you don't fire people. But anyway, I think I have to watch the talk to understand more. Number 5, he believes—

He doesn't fire anyone?

No.

Like the company doesn't or he doesn't fire?

I think he doesn't.

I was about to say, eventually, if you're running a company big enough, you have to fire someone. Yeah.

Number 5, he believes anybody can learn anything. He learned to be a CEO from being a— so he was a bathroom cleaner. He was a janitor before. I don't know if you know that.

The Nvidia guy.

Jensen. Yeah. Number 6, as he says, he tortures people into greatness. I think you and I can align with that.

Yeah, I can align with that.

So number 7. It's hardworking for me. Yeah, I think I think so as an executive or leader. What do you mean it's hardworking for you?

I expect the world from people and I'll pay them the world as well. Yeah.

Yeah. I mean, you do torture them into greatness.

Yes. I help them become better and I help them push themselves. It's really hard but I also take care of them really, really well. Not just financially but I help them out in whatever ways I can.

Yeah. It's no pressure, no diamond at the end of the day. I've learned this. You didn't have to learn this. You're just like this. But I've learned over the years that trying to be nice, trying to only just take care of people, that's not enough. You have to ultimately— it's the struggle, it's the push that helps them grow at the end of the day. Number 7, Stripe and AI are going to be massive. Okay, that's nothing. That doesn't tell me much. 8, Jensen thinks AGI is going to happen in the next 5 years. So artificial general intelligence, whatever that means to any of you listening. Number 9, Nvidia uses—

Let's go back really quickly. AGI, for anyone listening, it's like imagine AI going to the next level where it can pretty much operate like a human. That's what he's talking about.

It's basically better than human. Yes. Yeah. And a lot of people are debating, like Ray Kurzweil, he's a futurist. He thinks it's gonna happen in 2029. Sam Altman also thinks it's gonna happen in the next 5 years. People like Elon are very afraid of it, and I think rightfully so. We should think about ways to protect ourselves from it. But that's interesting, right? If we're gonna have human-level intelligence in the next 5 years, What does that mean for people in the working class, right? Which we'll talk about in a second. We should talk about AI agents right after this. Number 9, Nvidia uses AI heavily to create their GPUs. Number 10, one of Nvidia's core values is love and care, similar to Stripe's value of beauty and craft.

Huh.

Yeah. What's your main core value?

My main core value or the team's main core value?

NP Digital's main core value.

We have quite a few of them, like own it.

You can only pick one.

Okay, my favorite one is own it.

Yeah, you don't have a problem owning it. Ours is be improvement obsessed.

Yeah. But even though I don't have a problem with owning it, a lot of people do. Yeah, in the workplace. Yeah. And I believe if you did great, great. If you did bad, own it. What did you learn from it? Let's move on and give yourself a pat on the back for learning. And let's try to be better next time.

Most people don't like taking responsibility.

But here's the thing, if you own it, at least with me, I don't care if someone messed up. No, it's like, how do you learn from it so you don't make the same mistake over and over again? And if you mess up 20 times and they're all in different things, and eventually you learn what to do, great. Now you know what to do to produce results and make people happy.

You know, it's, it's, uh, I forgot which company this is with, but when someone made like a $250,000 mistake, at that time it was worth like a couple million dollars.

You're talking about IBM?

IBM. Yeah, IBM. It's like, it's— and the guy wanted to resign, right? I said, no, no, you didn't resign. I just, I just paid like a $250,000 for your education. You're going to figure it out later.

I don't think it was $250,000. I think it was way more than that.

No, $250,000 at the time. And then now in today's money, it's valued at a couple million. That's the story. All right everyone, quick message from the Agency Owners Association. So this is the peer group for agency owners that Neil and I both put together. This is for people that are doing 6 figures, 7 figures, 8 figures, even 9 figures as well. And we're all here to help you grow your agency faster. In this group, we share leads, we share learnings with each other. It's a community where people can ask questions, their most burning questions, Personally, professionally, we'll share templates, reports, things like that. We're constantly adding more value to the group. I will tell you that the price is continuing to increase. The good news is that there's no long-term commitment. So you can just learn more by going to marketingschool.io/agency. Once again, that's marketingschool.io/agency to learn more, and we hope to see you inside. So you and I have been talking on the phone about AI agents a little bit, and remember I talked about the tweet with Aaron Levy from Box?

Yeah.

Okay, just some context here. So Aaron Levy founded a company called Box, which is a storage storage SaaS. They compete with Dropbox. They're more known for enterprise storage. And obviously you have like Google Drive and all these now. But here's what he said. I'm not going to read the whole thing here, but in this tweet, Aaron Levy says, AI agents have the ability to fundamentally change the business model of enterprise software. Today, when you build a SaaS product, the primary business model is to sell seats that are tied to end-to-end users of your service. For same-store sales, you can essentially grow at the rate of the headcount growth of that customer or by adding additional capabilities on top of your offering. This business model, of course, is fantastic and has been around since the early days of enterprise software and will be going and will keep going forever. And then this is the last paragraph I'll read. But AI agents unlock an entirely new business model in software because you can now grow in a way that is uncorrelated to direct enterprise headcount or even external factors to your software. What's amazing about this model is that the AI vendor can actually, actually can directly grow itself within an enterprise account due to due to direct productivity gains. For instance, imagine an AI agent that can generate outbound sales demand for a startup. Traditionally, B2Bs will start hiring a very small team of maybe 1 to 2 people and increase resources in that area in a very incremental fashion. Essentially a process that looks like a guessing game where you try to predict how much demand that you're— there's going to be in the market. In a world of AI agents, you would simply set a budget and then decide how fast you want to grow. This is much closer to the business model of Google AdWords than it is a traditional SaaS service.

It's interesting. Curious to see how it actually plays out.

Yeah, I think it's going to make a major impact. So my thing is, in the next 5, 10 years or so, I don't think you're going to need as much headcount. These are my predictions, right? And you're going to have AI agents that can do the bulk of the work. And really, a lot of the people that you're going to have on your team are the top 5% of people at the end of the day. I told my team, earlier this week, I was like, there's this book called 21 Questions for the 21st Century written by the Sapiens author. And he wrote in that chapter, there's going to be a class of useless humans. I was like, guys, I don't know about you, I don't want to be useless. And I will not accept anybody at Single Grain wanting to be useless. We're going to embrace this change. Who knows what's going to happen? But we all know for sure AI agents are coming and it's already starting to happen.

So, so the way I look at it is the question is what industries does it impact the most and how long does it take corporations to adapt? And I'll give you a prime example of this. Okay, so at NP Digital, we work with mainly enterprise corporations and everyone talked about how the cloud is going to change everything, and it did. But dude, do you want to know how many publicly traded companies we work with that are still not fully on the cloud? It takes a lot of these big corporations years and years and years and years to adapt fully.

Well, still a lot of companies haven't even done digital transformation, right? It's taken 20 years or so, right?

So when people talk about these AI agents and stuff, like, don't get me wrong, in theory I agree, but it's not that simple to just integrate them within an organization. And I'll give you a great example of this. I interviewed this guy named Max. Years ago, he worked at Dentsu Aegis. It's a big ad agency based out of Japan, or I think headquartered out of London, but originally founded in Japan. Good-sized corporation, roughly $10-ish, $9, $10 billion in revenue. That's a good-sized business. That's a large corporation. Max's job was to implement AI and technology and automate as many things.

How long ago was this?

A few years ago.

How many years ago?

Few years, like 2, 2 and a half.

Okay.

To get rid of, um, to get rid of humans where it's not needed, right? I'm, I'm not really sugarcoating, I'm just being really blunt. I'm cutting to the point. Yeah, yeah.

You know, his cost—

yeah, his verbiage was much nicer how he explained it. You want to know what the biggest struggle was? To do that fully effectively and efficiently, he still needed somewhat help from the people in the original roles. If you tell someone, hey, I'm going to replace you with this AI, help me replace you, most people are like, screw off. And you want to guess how much traction he made?

Not much.

Very little. And does it mean that other companies won't get that? No, I think there's going to be a lot of companies that make a ton of progress with it. But I also think there's going to be a lot of companies that struggle to make tons of progress with it as well. And does it mean that those companies will fail or succeed because their competitors may be more efficient? I don't know. Time will tell because there's some companies that are happy to operate at 5, 10% margins versus 20%. Or maybe those other people cut down their costs, or maybe the technology doesn't work as good as it's supposed to. Like, do, do you remember Sam Altman talked about how it's scary on the type of content that, you know, AI can produce and the quality of it and how, you know, how crazy it is and hard it is to tell if it was written by a human or not?

Mm-hmm.

Do you remember when, yeah, it started coming out, ChatGPT and everything, and him talking about that? The content that AI produces today, and we produce a lot of it, like as in we test a lot of it, it's crap. And I'm not saying it won't get better, but in its current form, it's crap. I'll believe it when I see it. I know it is going to happen. Is it going to happen in 4 or 5 years? I don't know. Is these AI agents going to happen right away and be able to think like a human? And go really advanced. I eventually think it'll get there, but I don't think it's just going to be an instant switch, just like with the cloud. Oh, it's just going to replace everything that was off-premise. No, it doesn't. Companies just move so slow, and we work with so many enterprise companies, like big ones, right, that are publicly traded, worth hundreds and hundreds of billions of dollars, if not trillions. And crap just moves slow, no matter what people say. Like, technology may be there, but they still move slow. I think startups will embrace it really quickly, and it'll change the way startups do businesses. But not necessarily the way that big corporations do businesses or business right away. Eventually, sure, but not right away.

All right. So we're going to come back to this monologue in 5, 10 years or so. I disagree with Neil. And what I will say is that the way cloud— not really much, right? When you look at the internet, big step change. When you look at AI—

No, cloud was a huge—

You can't interrupt me. You got to let me go. So, you went monologue like 5 minutes, my turn.

Okay.

So, cloud, not as big of a deal as the internet, right? Not as big of a deal as electricity, not as big a deal as AI. When you have some of the smartest investors in the world going like, this is probably bigger than the internet itself. Like, you know, you can look at the biggest tech founders out there. And the difference this time is that the Microsofts of the world, the Googles of the world, all these people, they're deploying a ton of assets into this, right? The cloud, like, yeah, the cloud was big, but like minuscule to me. In terms of how big AI is going to be, right? Or AI is right now. And to your point, which you just reinforced, you said that you guys are producing a crap ton of content with AI right now. And I do agree, like, a lot of stuff is— there's a lot of stuff left to be desired. It's not that good right now, right? But keep in mind that this started to really get bigger in the last year and a half or so. So it's like, it's still a fart in the wind right now, right? But the fact of the matter is there's a lot of resources going into this. I just don't see a world in the next— I do agree it's going to take time for stuff to change overall. I think we can agree on that part. But I do think a lot of people are going to be at risk because the things that we're seeing right now that we're building with Copilots and we're reviewing code internally, we're just moving a lot faster. Imagine someone like— imagine you 10 years ago, no MP Digital. Imagine me 10 years ago. We have a lot of ideas. All the stuff that we want to build, In the next 5, 10 years or so, these things are just going to build it for us, right? Am I saying all humans are going to be replaced? No. I'm just saying that the best of the best, those people will always have a job and potentially other people are going to be at risk.

So I actually think we're on agreement. I'm just getting at in which I don't think it's going to be 5 years where it impacts most big corporations.

I think— okay, I agree on that. Yeah. I didn't know you were just focusing on just the big corporations.

Yeah, I do think it's going to have a huge impact on a lot of jobs. But the transition period between startups and big corporations is going to be drastically different. You're going to see startups out there from day one leveraging all this technology and adapting. And big corporations just have way too many layers and bureaucracies. And a lot of them are going to adapt, but they're going to adapt much slower. And you're going to even see some not adapting. And who knows what's going to happen?

Okay. So I would say who's, who's the big tech company that's moving really fast right now with AI stuff? Microsoft, right?

Yeah.

Microsoft is. I would say Google is, but they're kind of like stumbling a little bit, right? And I can't remember the last time someone's like— for example, we're talking about advertising for this podcast, right? And Microsoft is going hardcore on AI ads across the board. They're trying to push their Copilots everywhere, right? And so they're going hard. They're deploying a ton of resources towards it. And I don't think I've seen a time when all these big tech companies have all rallied together to do this. I don't remember the last time. Do you?

I haven't seen this in a long time.

Yeah.

But the big—

Or ever, ever.

Well, they did rally around the cloud altogether. They pushed out—

Okay, go on. Tell more about the cloud.

No, no, but in which all tech companies were pushing for the cloud and they were talking about it as the future. You're seeing the same rally on AI. I do agree AI is going to be bigger than the internet, right? There's no doubt on that. I agree with you as well. I don't think cloud was as big as the internet. It's just more so the time for people to adapt is going to range drastically depending on the company size and the bureaucracy that they have. And like, for example, I know a lot of people that work at Microsoft, dude, they're making leaps and bounds in investment in AI.

Yeah.

But if you talk to a lot of people who work at Microsoft, and I know tons of executives, there's still tons of divisions and departments that are backwards and running like they should have been running You know, are running like they're like companies that are ten years old, right? Like they're really behind. It really depends on the division, the department. It's hard to make a company of that size adapt and be perfect for everything and for today's world or the future of technology.

All right, so I wanted to take a second to tell you about leveling up founders. This is an event slash mastermind for people doing seven. 8, 9 figures. These are founders that are doing that amount. And we've been doing this event for a couple years now. We've had amazing speakers, but more than anything, it's about the people. What I mean by that is like-minded people want to hang out with like-minded people, and this is the best spot to connect with people of a certain caliber. And we vet every single person that comes through. So if people are— even though they make a lot of money, but their character, we don't really align with that, we're not necessarily going to let them in. We do this event in Beverly Hills. And it's happening in the beginning of August. All you have to do to learn more about it is just go to levelingup.com/founders. Again, it's levelingup.com/founders. If you want to hang out with amazing people, Neil, my podcast co-host, is going to be there. I'm going to be there. People like Syed Balkhi, he will be there as well. And again, it's going to be a great time. levelingup.com/founders to learn more. And we'll see you inside. Dude, check this out. I'll try to link to this graph, but I think what we should get at too, to show people is adoption curves, right? So when you look at technology adoption curves, so you have AI in the red over there.

Is that the McKinsey chart?

No, this is eMarketer actually. So we don't know how legit the source is, right? So eMarketer, basically generative AI from 2022 to 2025 is that it went from 7.8 million users to about 117 million users. Tablets went from 12.9 million users to $124 million by year 4. So it's got a steeper adoption curve, it's growing faster, but by year 4, tablets are winning out. But I think it'll pick up again. Yeah.

And look, I, you know, time will tell. But in the next 5 years, you're gonna see a lot of startups come and they're going to be leveraging AI technology, and they're going to disrupt a lot of corporations. I think that's going to be a huge wave of potential investment opportunity is companies who are just more efficient doing what big companies are doing, but they're just Doing it way better because of AI.

Do you know how long it took for the average American to use electricity from its inception?

No.

So, from about 1900— 1910, sorry. 1910, electricity comes out and maybe there's like 5% of the population using it. By the end of 1950, it's 100%. So, it took 40 years to get there, right? But the adoption curves get steeper over time. It grows a lot faster.

So, that's with everything. Right now, when you're talking about AI, dude, people in this world, there's more people with cell phones than actual laptops and computers.

Yeah, smartphone adoption from like 2005 until like 2020, it goes up to 75%. Yeah, yeah. So food for thought, everyone.

You know, dude, it was funny. I was on a conversation the other day with someone who ran Reliance Jio, specifically the cell phone network.

Very big company.

We met him in Dubai. Do you remember that? Oh, yeah, yeah, yeah, yeah.

Because at my— at the dinner yesterday. Yeah, yeah, yeah. So You met with him when?

No, phone calls.

Zoom.

Okay. Yeah. Okay. So he was going, funny enough, at the airport. He was heading to the airport from Mumbai, I think, to Dubai. And we were just having a phone call and catching up. So when he was at Reliance, the big thing that he ended up doing was just empowering Indians with data in India, right? And then eventually Microsoft and I think Facebook invested. It was either Microsoft or Facebook or both. Took a cut of the company. And the whole model was, we're going to make it easy and pretty much really affordable for you to send texts, browse the internet, etc. The cell phone market just skyrocketed after that because then people could leverage technology. And it also changed India's economy.

They subsidized the—

At the beginning.

Yeah.

Then eventually they increased the price, kind of like the Microsoft Teams model.

You know, it's like, once they lock you in, you're good. You're stuck.

Yeah. And what he was getting at is like with AI and everything that's changing, They're seeing mass adoption in India. Why? Because everyone has these smartphones and devices where it's easy to use it. Back in the day when stuff got published on the internet, well, not everyone had the device to actually log onto the internet. Yeah. My first modem was AA batteries and you would plug it into the wall. And if someone called, you get a screen. Batteries in your modem?

Yeah.

I don't remember that.

I always had like a card that I stuck into my computer.

So you had it before than most people. You had a card in your computer.

Yeah. Like a 288 and 56.

But then there was another thing that plugged in from your telephone line. With the battery and then into the computer so you can have internet.

I don't remember that.

Yeah.

Yeah. I just remember I had an extension cord going into the phone line.

Uh, yes. Going to the phone line. Yeah. Eventually it got switched to that, but before there was something that required batteries and it was an external device and then it became the cards. And when someone called you on your phone line, it would disconnect you from the internet, which sucked.

Yeah. I don't know about you. I stole my mom's credit card to sign up for the internet. She wouldn't let me do it. And then I would wait for them to leave and then I would move my computer desk because it had wheels over. Because I didn't have the extension cord, so I'd move it next to the phone thing. And then once they came home, I'd like unplug it and I'd move it all the way back. Yeah.

It worked out well for you.

It worked out. Yeah. By the way, I had something else to say about— I had something else to say about AI. It just escaped my mind. It'll come back.

But speaking of AI, we're still getting a lot of companies, we're seeing it more than ever before, of pushing us, not allowing us to use AI in our work.

For the content that you generate?

For any of the work.

For like, let's say you're a publicly traded company and you're worth $200 billion. You just don't want the risk.

They don't want the risk.

Yeah, yeah, yeah. No, not you.

Yeah, they, they don't want the risk. We're actually seeing it more and more frequently.

That makes sense. Seeing that. Yep. Because you're a big company, you got to de-risk.

You got to de-risk.

Yeah.

All right.

So that is it for today. Go to marketingschool.io/agency to sign up for our Agency Owners Association. The group is continuing to grow. And by the way, we're upping the price every now and then as we continue to increase the value there. So go check it out if you want to grow your agency faster. Marketingschool.io/agency. Don't forget to rate, view, subscribe. It helps us grow. And thank you for your attention. Check us out on YouTube as well.