Why Intelligence Pales In Comparison To This

Neil and Eric break down why high agency is becoming more valuable than raw intelligence in the age of AI. They discuss Andre Karpathy’s views, Klarna’s AI experiment, why speed and execution now beat perfection, and how build versus buy decisions are changing. The conversation covers AI-driven productivity, the future of agencies, founder-led growth, and why adapting early is critical for entrepreneurs, marketers, and operators heading into 2026 and beyond. Key Takeaways: High agency beats intelligence in an AI-driven world Speed and execution matter more than perfection AI is reshaping agencies, startups, and big companies Chapters: (00:00) Agency vs intelligence explained (00:30) Why high agency wins with AI (01:12) Klarna, Karpathy, and AI impact (03:00) End of large organizations? (04:10) Build vs buy debate (05:46) Speed, execution, and founders (07:58) AI tools and productivity gains (10:30) AI, GDP growth, and markets (13:05) Why 2026 looks bullish

• 0:00Speaker 00

  Finding customer information shouldn't mean digging through emails, call notes, and spreadsheets. HubSpot's starter plan brings all of those details together, so it's easier to track what's happening with every deal and what to do next. It pulls information from calls and emails. Then it gets to work, logging and summarizing them. Right after your call ends, HubSpot pulls next steps from your call notes and drafts suggested follow-ups. And it stays up to date for you, so you can build demand, win more deals, and spend less time on repetitive work. Your records, insights, and next steps are ready for you to act on. And you don't have to miss a beat searching for any of it. Try the starter plan for $10 a month. Learn more at hubspot.com. Neil, if you had to pick between agency and intelligence, what do you think is more important? And do you want me to describe what agency means?

• 0:40Speaker 01

  Agency is a company that just goes, assuming you're talking about marketing agencies, but a company that, just agencies in general?

• 0:48Speaker 00

  No, I'm referring to the word agency, like a trait. And so you have intelligence or agency. So when I, when I say agency, someone who is high agency is someone who knows how to figure things out. They know how to get shit done, okay? So in today's day and age, what do you think is more important? High intellect or high agency?

• 1:04Speaker 01

  High agency of just getting stuff done, because you can get a lot of the intellect from AI.

• 1:09Speaker 00

  Yeah, okay, so check this out over here.

• 1:11Speaker 01

  I do need both, though.

• 1:12Speaker 00

  Yes, yes, hopefully you would have both. But Sebastian Siemtalski,

• 1:21Speaker 00

  this is Klarna's, Klarna's CEO, that's what I was just saying, okay? So remember Klarna back last year, they were going all in on their agents? And then they kind of had to roll that back. So he tweets this, 2026 will be the-

• 1:34Speaker 01

  You're talking about Klarna, the buy now, pay later company?

• 1:37Speaker 00

  Yes, correct.

• 1:38Speaker 01

  CEO is like, hey, we got rid of a lot of our support people. But then I think he went back on that and started hiring them.

• 1:43Speaker 00

  Yeah, he rehired them. So anyway, he tweets this, 2026 will be the year when AI starts affecting for real. Been preparing for this since 2022. Software dev costs dropping to zero. Context engineering will be central. Build will win over buy, which is something we can talk about. Build will win over buy, I want to highlight that one for a second. In large organizations, doing yourself, doing things faster yourself versus delegating, uh, agency now supersedes experience and know-how. And this is a reply, it's a quote tweet. Based on Andre Karpathy's from yesterday or two days ago, saying that he's never felt this behind his programmer, because there's all this new stuff that's coming out all the time, subagents, MCPs, like all these other things, right? He's like, I have a sense that I could become 10x more powerful if I just properly string together what has become like, this is one of the top guys in AI.

• 2:28Speaker 00

  And so I think this is interesting because Andre Karpathy earlier this year had, has basically said that agency is greater than intelligence. And so what do you think about this? We can, we can go ahead and react to this because there's a lot of implications here.

• 2:41Speaker 01

  I think you ideally need both. I don't think it's one or the other. I think, um, forget 2026, if you want to go forward a few more years, if you don't have both, I think you're screwed. I think you're already screwed right now. Like if you look at the top people who get paid the most, they'll have both.

• 2:58Speaker 00

  Yeah. You know, we talked about this a few days ago, but when I was talking to, to co-founder Sean, he's like, yeah, well, I basically generated close to a billion tokens. I'm not saying that's a good proxy. He's also, he also, he's also written 10x more lines of code than he has usually. And we're, I think now when I use AI studio or when I use all these things, we're just able to do a lot more with, with less. And this is an interesting comment over here. This guy class tweets this, he says, wouldn't this also essentially mean the, the end of large organizations? So he's like, and then Sebastian responds, he says, yes, it very, it very well might. Two trends will meet large org CEO doing founder AI mode, small companies growing insanely fast.

• 3:42Speaker 00

  They meet up in the middle and we're already seeing this very quickly. There's a lot of companies getting to 20 million ARR in like three to six months or so, hundreds of millions in ARR in, in, you know, one to two years. And I, I think it's, it's tough if you're like, uh, if you're incumbent, if you're a very large like an agency holding company, like what do you do in that situation? I think the biggest thing you can do right now, if you're listening to this and you're 18, 20 years old, or whatever it is exactly, is you go out and you experiment with these things. This go back, this goes back to what I was saying. I'd like, if you say you don't have time to learn this, even spending 10 to 20% of your time, where you're working hours, learning this stuff, um, it's crazy to me to not know this and you're going to have such an advantage over other people who have been in the Workforce for a very long time who don't want to adapt and think they can get away with this.

• 4:24Speaker 00

  Um, the reality is they can't because again, my conversation with with um, you know, um, uh, the CEO of Zapier. He's like, yeah, we do hackathons every three to four months or so. Um, we shut, initially they shut the company off or or or down for for a week and they just focus on getting everyone's AI adoption up. Um, and this is a baseline requirement now. And so agencies, number one, um, I think this other, this other comment over here, Neil's interesting. He's saying build will win over buy. What do you think about that?

• 4:51Speaker 01

  I think it depends on the context. And when I say context, um, a lot of times people buy companies because of their attention and audience, not because of the product or the service. A lot of it's the brand and the recognition they've built up over years, which is hard to replace from just building. Again, I always think there's context, right? In some cases, yes, some cases, no. I think for a lot of stuff like we bought a a writing company. Uh, I don't know when, let's call it a month ago.

• 5:23Speaker 00

  Yeah.

• 5:23Speaker 01

  We did it because it'll take our dev time down by more than half. I think they actually said, we'll end up getting out an MVP version. We bought it, I think sometime December, we'll get it out by the end of January. Let's call it, give my team another week or two in case it delays. But if we didn't do that, it would have taken us a few more months on top of that. And then money we spent was cheaper than the extra time it would have cost to pay engineers.

• 5:46Speaker 00

  This is the beautiful thing. Like, I, I think your team, um, and you're, maybe you're not in the details on this, but we're not even going to be saying the word MVP anymore. It's just going to be like, hey, just get a prototype out. On a prototype now is just like a vibe coded thing that's designed well, or you're using something like alloy.app and it takes your screen, um, you take a screenshot of something that you like. Neil, we, we used to do this all the time. We'll screenshot things that we like and say, hey, can you go to your dev team, can you do this? Or your designers, can you just do this? Now it's just like, you just

• 6:10Speaker 01

  Yeah.

• 6:11Speaker 00

  Take the screenshot, drag it over to Alloy and say, hey, just fold it into my app and boom, you have a screenshot and then you just move a lot faster. That's where we're going, guys. Things are just going to move a lot faster.

• 6:20Speaker 01

  Yes. And and I think that's really a big key in winning, like whether it's entrepreneurship or marketing. Really, speed and execution matter so much. People stress about perfection. Perfection is overrated. I'm a big believer in executing and being directionally correct. You don't have to be correct all the time, just be directionally correct and just keep firing in that direction. Just keep pushing hard and go through as many walls as you need. That's how I tend to operate as a founder. That's not necessarily the most efficient way, but it's worked for me.

• 6:51Speaker 00

  I actually think that's the only way because there's a, there's a book that I have behind me here called Why Greatness Cannot Be Planned. And when you think about this going into human psych or human evolution over time, right? So we started out as a single cell of, of bacteria and we became what we are now. But who knew this was going to happen? Or someone that invented electricity, who knew that it was, it would power all these things that we use today, right? And so a, all that to say is. Frictionless Growth Marketing hosted by Sonya Thompson is brought to you by the HubSpot Podcast Network, the audio destination for business professionals. Frictionless Growth Marketing hosted by Sonya Thompson shows how modern brands grow by removing the friction in their marketing and customer experience that pushes customers away every single day, often without leaders even realizing it. Each episode delivers practical insights, real brand examples and conversations with marketing leaders,

• 7:36Speaker 00

  showing brands how to win with the new rules of growth, identity-driven marketing, cultural relevance, and frictionless customer experiences. Listen to Frictionless Growth Marketing wherever you get your podcasts. If you want to build something that's great for yourself, you just need to take one step forward every single day. Imagine yourself, you're in a marsh, okay? And you only have one stepping stone in front of you. You're going to take one step forward. Sometimes you have to take two steps back. And that's all it is. There's, there's like, everything in front of you is, is fog. You just don't know what's going to happen. So all that to say is high agency, high speed. And by the way, high agency people have high speed anyway, right? Yeah. So here, let's, let's come over to this real quick. This is similar web traffic.

• 8:18Speaker 00

  And we see, okay, a month ago, ChatGPT, okay, let me, let me go back a year. 12 months ago, ChatGPT was 87% of Gen AI website traffic share. Gemini was 5.4%. You fast forward to December 5th, ChatGPT is 68% and Gemini is 18.2%. And you know what's interesting, Neil? Last week, we talked about how you said that your nephews and. Maybe the young kids, they're all using Gemini. So I actually tweeted this out and it's it's actually a mixed bag. Some people are using, some kids are using perplexity. Some kids are using chat GPT. A lot are using Gemini. It's it's kind of a mixed bag right now. And the reason they say chat GPT is for old people. It is because old people like us keep talking about it all the time.

• 8:58Speaker 01

  No, but I I've talked to enough young people after that. And I still say the majority end up using Google products and the reason being is not because it's better or worse, it's because it's free. They don't have credit cards to spend. Can't go to your parents and be like, I want to cheat on my school where can you pay for a upgraded version of chat GPT? It doesn't work.

• 9:16Speaker 00

  Yeah, I mean, sixty eight percent, though, to drop from eighty seven point two percent to sixty eight percent is pretty significant.

• 9:21Speaker 01

  I think, and here's what people aren't taking into account. Google through their AI drives a lot more than that in traffic. A lot of it's through AI reviews, which I consider AI, right? But that's not included in this.

• 9:34Speaker 00

  Here's here's a here's a fun one. So.

• 9:36Speaker 01

  I bet you if you put AI reviews, Google would be driving more traffic than chat GPT.

• 9:40Speaker 00

  I think that's kind of cheating, though. It's like when when WhatsApp or was saying or or Meta was saying, oh, we have so much usage and but it's like that that cheap little like search bar.

• 9:49Speaker 01

  My favorite is on Meta. We cross a hundred million users on threads so fast. I think they beat everyone in a record speed. Because they just, but who the heck uses threads?

• 9:59Speaker 00

  Threads is a cesspool. I told you. So so I I enjoy my time just trolling people on threads because it gets a lot of engagement. So whatever I'm feeling like I I'm feeling feisty, I just go on there and just like, you know, click away on my keyboard.

• 10:13Speaker 01

  Dude, if you haven't seen Eric on threads yet, it is entertaining. Just go log on threads, go look at Eric's posts and then go look at the comments. Go look at the ones that were really taken off. They're really polarizing viewpoints. I don't think they're too extreme or anything like that, but the comments, like just some of the people. You could just tell where people stand politically on threads is very, very clear. It's also pretty clear on X as well. They're both very.

• 10:41Speaker 00

  Well, here here's what I'll say about these two platforms, right? And not to be political or anything, it's. Threads vilify success and I'm just like all the time you spend vilifying success you could spend time making yourself better. What I use, my X is more finely tuned and so I'm constantly learning. There's AI stuff I see popping out, the latest research, all these papers, right? I'm dumping into no big, it's very productive for me. And so I'm almost trying to be like the, the old uncle on threads and just trying to educate people and turn people around to, hey, spend time investing in yourself. Don't spend time pointing, but I'm, I'm very, uh, I'm very sassy about the way I do it. And so I'll just leave it at that. So, um, Neil, speaking of X over here, what do you think about this? So Elon Musk says this, okay, double digit growth,

• 11:23Speaker 00

  okay, we're talking about GDP growth is coming within 12 to 18 months. If applied intelligence is proxy for economic growth, which it should be, triple digit is possible in five years. I need you see this from rounders, Matt Damon going all in.

• 11:37Speaker 01

  I, I do believe that the economy is, is ripe for a bull run.

• 11:41Speaker 00

  Look at this, Marc Andreessen, it's not even time to build anymore, Neil. It's time to grow. And then after hearing Scott Bestin, I can't help but feel bullish. But the thing is when everyone's feeling bullish, maybe that's a time for a correction. Um, but when you look into, you listen to Scott Bestin, you listen to David Sacks, it's like, it's like dude, um, even Scott Bestin saying like, I saw someone tweet, he's like, he's saying, hey, guys, you know, money's gonna fall from the sky in 2026. Uh, but yeah, 4.3% year over year GDP growth in Q3 is kinda effing crazy. Didn't see that coming.

• 12:10Speaker 01

  Yeah, but they're, they're saying the growth is gonna be even more next year.

• 12:14Speaker 00

  I, I believe it.

• 12:14Speaker 01

  I don't know, seeing and baking dude, so many of my friends who have businesses are getting bigger loans, more flexible terms and cheaper rates than ever before. I'm talking about in the last five, 10 years, I've never seen anything like this, but it's because of deregulation. It's just the average person does not see it. Like if I go to a bank right now and asking for money, it is ridiculously easy and cheap for almost anything right now, even in this economy. Like banks are throwing money on anything. I got someone hit me up the other day. They're like, you don't pull money out of your house? They're like, you can pull up to 75, 80% interest only. And I would have a jumbo mortgage and the jumbo mortgage isn't covered by what Fanny

• 12:57Speaker 01

  Mae or Freddie Max would be the bank using their own money. They're like, we'll be really flexible. We'll give you better rates. I think they offered me, it was like 4.0, you know, like six or something like that percent, which isn't bad for, um, an interest only loan on a jumbo product. I didn't take it.

• 13:15Speaker 00

  Here here's what I'll end with. Actually, you brought up a good point. Okay, go ahead. Finish your point.

• 13:18Speaker 01

  And and by the way, you know, as you start making more money, uh, the financial terms get even better. You know, like I have a mortgage on one of my homes. Okay. You know, if I want to change my rate, I don't ever have to refinance. I just call them up and say, can you just change me to the most recent rate? I think they charge me like a thousand five hundred dollars or a thousand two hundred fifty dollars each time. And they just update my rate, very little paperwork. When I say very little paperwork, they just sent me a one pager or two pages. I think it's actually two pages, sign and just mail it off to them and, you know, send them the digital copy that I just authorized them to charge me a thousand two hundred fifty bucks.

• 13:53Speaker 00

  Yeah. So if if you guys can't sense it, Neil and I are feeling a little bullish about twenty twenty six. We feel we did feel bullish about twenty twenty five, but I think there are some, you know, stuff to kind of get set with this foundation. Um, but you know, Tom Lee, you know, the the fun strat dude, right, that you see on CNBC all the time. Um, so he actually thinks Goldman Sachs and JP Morgan will actually hit mag seven status, right? And I was like, what? Um, so his reasoning is basically on the transfer transformative impact of AI and blockchain. You're gonna see a lot more stablecoin news. You're you're talking about credit terms loosening up, interest rates are going down. Scott Besson was saying there's gonna be a lot more loans, which is what you just mentioned. Um, the impact of AI as well, they're probably gonna, you know, probably they're gonna become a lot more profitable, a lot more efficient. And all the money that they have, um, all the lending that they're gonna be doing.

• 14:35Speaker 00

  And so he's just, I I think he's just saying that, um, you want to take a look at Goldman and JP Morgan. So not financial advice, but I I think it was an interesting thesis.

• 14:43Speaker 01

  It is. And, uh, when you look at the lending side, which really helps make the economy go round and round. I'm getting a lot of banks hitting me up and I have been for like six, seven months now being like, hey, do any of your friends need loans? Like, no joke, banks are asking me for introductions because they're like, our bank's pretty flexible. They have a really strong appetite to lend more money right now. And like, I've been meeting up, I I did lunch with, uh, JP Morgan, uh, um, the Mayborn here.

• 15:13Speaker 00

  What are you wearing your watch today?

• 15:15Speaker 01

  The what?

• 15:16Speaker 00

  Is it why you're wearing the watch today?

• 15:17Speaker 01

  No, I usually wear a watch when I go out of the house.

• 15:19Speaker 00

  Oh, you do?

• 15:20Speaker 01

  Almost always. It's when I go downstairs. I consider that part of the house, so I don't wear a watch usually. But if I'm wearing a watch, that means I went out of the house. Uh, JP Morgan, um, Truist took me out as well. Um, and there's been a few other banks that I've talked to. Keybank. Um, Zion's bank has been hitting us up, looking for more loans and people. But like, I'm telling you, the, the number of companies these guys are all looking to fund, they can't find enough of it right now. And they're all telling me the same thing. Our management is telling us, hey, go find more loans. We will loan right now. They're trying to be super aggressive on rates and everything.

• 15:55Speaker 00

  Yeah. Knock on wood. Here, here's the last thing I'll read. I got this from Grok. That's what we're talking here. So, um, Lee highlights how these technologies will dramatically reduce quote unquote employee intensity in financial services, allowing firms like Goldman to operate with fewer human resources per dollar revenue. This efficiency gain is expected to drive substantial margin expansion without necessarily requiring top line growth as AI automates back office tasks, risk management and operations, while blockchain streamlined settlements, tokenizes assets and enhances payment systems. As a result, banks could shift from being valued as traditional lenders with lower multiples to high growth, software-driven platforms, leading to a two to three X rerating evaluations and trading more like hyper-efficient tech companies. We'll see. I can see that happening.

• 16:37Speaker 01

  I hope so.

• 16:38Speaker 00

  Yeah.

• 16:38Speaker 01

  Time to tell.

• 16:39Speaker 00

  All right, guys. Uh, that is it for today. Please don't forget to rate, review, subscribe.