Is AI a Bubble Right Now?

Do you believe that AI is in a bubble right now? More specifically, AI stocks. So when you look at like Nvidia, for example, or you look at any of these other tech stocks, do you believe it's in a bubble? Do you— or do you believe it's more— do you think it's like 1999 or what do you think it's like?

Are you talking about the valuations or the revenue?

Both, actually. So let's, let's, let's, let's look at that because people tend to look at valuations first. But I like to look at both valuation and revenue, right? So like if we look at it in aggregate, how do you, how do you look at it right now?

So I don't know if valuations are in a bubble or not.

Mm-hmm.

What I look at the NVIDIAs are is they're the shovels, the picks, the axe of the gold rush. Do I think there is a lot of fluff in AI and companies that are going to fail? Yes. Do I think the energy companies, the NVIDIAs and stuff like that are going to get crushed? No, I think they're going to keep generating more revenue for a very long time. Does that mean that their stock is going to go up or down? That part I don't know, but they have so much demand because Think about how much AI can power from helping with cancer research and new drugs and formulas and all this kind of stuff. It has solid use case. The problem I'm seeing in AI, and this is what some people could assume is a bubble, and what's a 24-hour VC, Harry something?

20VC.

20VC, what's his name? Harry—

Harry Stebbings.

Stebbings, yeah. And he talked about how the valuations are just becoming ridiculous. And he posted something on X, I think a few weeks ago, We're seeing a lot of crap companies generate tons of revenue. I'm not generating tons of revenue, get a really high valuation when they shouldn't be.

Mm-hmm.

A perfect example of this was I was talking to an AI company in the video space. They do well into the 8 figures in revenue and they're talking about how, you know, their company is worth, call it half a billion-ish dollars. And I looked at it and I was like, Google has a better product. And Google's product is free to cheaper, right?

Yep.

And I was like, I don't understand why people need to use your product when Google's product runs circles around you. And oh, by the way, they can make it free. And they're like, no. I'm like, yes. If that video product helps advertisers spend more money on Google Ads and they go from, call it, $300 billion in ad revenue to $350, They don't mind losing $1 billion to give away that product for free completely to generate $50 more billion in ad revenue. And that's where I see a bubble in which a lot of companies like the Jaspers of the world were funded with high valuations when it never really made sense because it's what? So OpenAI creates an API, you can do this on ChatGPT. Sure, Jasper is creating some unique features, but—

Yeah.

You can do most of it with the open API.

Jasper was interesting back in the day before ChatGPT came out because it was ChatGPT-esque, right? And that's where, that's where they were accruing the value. And then, well, what happened? Not so much anymore. They actually still have a lot of employees.

You see that?

I checked on LinkedIn.

Yeah, the ex-Dropbox president Tim Young is the CEO and runs it. I don't know what their revenue is, but I've read something online at $50, $60, whatever million dollars. Don't get me wrong, great business. Yeah, but SaaS companies are not valued at $1 billion plus.

They're a multibillion-dollar valuation.

I know it was more than a billion. I don't know if it was multibillion, but I know it was more than a billion.

I think it was like 2. That's what I read.

Yeah, I think it was 1 and change.

You could be right. I remember 2 or 1.5. If you want to, you're compelled to search it. Okay, let me give you my take on this. So the reason why I brought this topic up, so is AI a bubble right now? I don't believe AI is in a bubble. Go ahead.

1.5.

1.5.

See?

Okay, so is AI in a bubble? I don't believe so. And here's the reason, because If we go back to the concept of Jevons paradox, right, you think about the electric— the electricity, electricity back in the day. You think about the internet back in the day when you and I, when our internet speeds started to get faster and faster, what started to happen? We started using it more and more and more. When we're getting more and more tokens now, really what we're buying is intelligence, the cost of intelligence, right? The cost for us to code things like we just mentioned, the efficiencies we're seeing with our engineers, our developers, right? So I'm literally arguing with a finance YouTuber right now on this, and I bet him $10,000 that— here's my bet. You can tell me if my bet's unreasonable, but—

Okay.

I basically said, look, you look at the usage, it's going up right now. Companies are literally cutting headcount. Not saying that's a good thing, but they are cutting headcount. That means they're seeing value somewhere, right? If Accenture is going to cut 11,000, Salesforce cut 4,000, and Walmart's like, oh, we're going to keep our headcount flat. Okay, something's happening, right? So if he says, okay, well, you know, Nvidia, this is a— he's talking about the round tripping with the revenue where like, you know, they're investing $100 billion in OpenAI and then like $300 billion OpenAI is paying to Oracle and like they're round tripping this revenue, right? And he's saying it's more than this. This is like Enron, like red flag type of stuff. I'm like, dude, like OpenAI has 800 million users. And he's like, well, the $20 a month is never gonna grow into the revenue that they— I'm like, dude, in startups you go for the users and then you can turn on ads later. You can do whatever you want with the users later. It's not just gonna be a $20 a month thing, right? And he's arguing with me back. I'm like, okay, I'll bet you $10,000, okay? I bet you OpenAI Anthropic Cognition in the next 5 years, I'll even give you 10 years on it, their valuation plus their valuation or revenue is going to be higher, okay? And if you want to look at the Mag 7, he was using Nvidia. He's like, why do you think Warren Buffett doesn't use Nvidia? You talk about Warren Buffett all the time, right? I'm like, yeah, I'll tell you what, Warren Buffett doesn't invest in tech stocks. He's tech stocks.

I'm like— Some he does. He has Apple.

I think— sorry, I said that. I said Warren Buffett tends to avoid tech. And so he does have Apple, he does have Google. But guess what? It's his lieutenants that put in that money. He has said in his circle of competence that he doesn't touch that stuff. Right. And but you and I both understand technology and we know how this stuff goes in the long Long run. So I was like, look, okay, you can take Nvidia, you can take Microsoft, even take Amazon as well. You can take Google. And I'm telling you, in the next 5 years, revenue valuation will be up.

Okay.

And I'll bet you $10,000. I'll bet you more than that.

How many years? 5 or 10?

I can do 5. I said 5 first.

I said even 10 is fine. I agree. I think within 5 years they're all worth more. Yeah, but this is not financial advice from Eric and I. This is just our own opinions. We're not financial experts or economists and you could lose money. But I agree with you. I think his comment on Enron is very non-relatable to what's happening with like Nvidia investing. Enron was taking revenue and let's say the revenue was over 3 or 4 years. They were booking it all right now, which you're not supposed to do in accounting. If someone pays you over 4 years, you can't just count all the revenue today.

Yep.

And that's where Enron got into a big issue. If Nvidia is making money from selling GPUs and OpenAI is trying to build massive data centers, and you know that company's growing in revenue, these companies have venture arms. Why can't they invest in 'em? It's like, it's taking money off your balance sheet.

Yep.

Yes, they're spending some of that cash and investing it back into you, but you end up becoming an owner in that organization that's growing very large.

Yep.

I look at OpenAI and Nvidia in a world where there wouldn't be— where a world where monopolies would be allowed, you would start seeing these companies gobble each other up because they're so dependent and you would grow even faster. But we can't do that in today's world.

You know what Jensen said? So I was listening to a podcast where he was talking to Brad Gerstner, right? They're doing an interview. And then one of his partners was there too. And so they brought it up. It's like, what about this round tripping? He's like, dude, like, I believe that OpenAI— Jensen saying this— that OpenAI is going to be the next multi-trillion dollar company of our generation. I can see that. Right? And, and he's like, we regret not putting in more money because we were poor, right, before everything took off, right? And now here's a chance for us to put in more money, and we're gonna put in more money. And that was his reasoning. And like, it actually goes to what you said, you— I'm sure you didn't listen to that podcast. That's just like, it's the same logic.

It's the same logic. Yeah, it's like Google invested in Uber early on, right? Or not that early, but early enough. And Uber started using Google Maps, right? So what, Google's sending them money, then they're making money because they're using Google Maps? I don't know what the deal or the terms of the deal are for them to use it. You could say it's going in a roundabout way, but Uber's map technology sucked compared to Google's. They use Google's, it became a better product.

Yeah, they—

it's a win-win scenario for both of them.

I look at this as just the, to kind of wrap it up. I think it's, uh, Look, you— there's real revenue, there's real products here. And then I think if there's any investing lesson, not advice for you guys, it's Warren Buffett himself. Uncle Warren Buffett has always said you want to invest within your circle of competence. And the beauty with investing is you can wait to swing on all the pitches you want and you can wait for the fat pitch and then you can swing, right? And so— and he has said like, tech is not my circle of competence. And so he doesn't bet there. And so— but when you know this, this area, you can afford to bet more than— not financial advice, right? But when you understand your niche You can press hard.

Yes. But no. Yeah. And you're right. He tends not to invest in tech. It's his team members that do. And it works out for them because they understand it. And you and I are similar. We only invest in what we understand, whether that's right or wrong.

Yeah. Well, I mean, you and I have been stupid with angel investments, but yeah.

Angel investing has worked out for me.

It has?

Yeah, it worked out really well.

I would say I— net-net, it's worked out, but I feel stupid because most of the bets don't work out. That's why I say it's stupid.

So I stopped I stopped doing angel investments. I don't know how many years ago, but it's, it's been a while. And then I shifted to just being an LP in funds. So when I got started as an angel, companies never really raised more than $100 million in total. I'm not talking about one round, I'm talking about in total. And they would just go public if you, you know, you raised $10, $5—

This is a long time ago.

Yeah. $1 million, your Series A, $4 or $5 million, your Series B, you know, and then you got to D and then you've raised maybe $100 total, you go public. And the returns were actually really decent if you picked the right ones. The problem that I started facing towards the latter part of my angel investing quote unquote career, but it wasn't really a career, more so a hobby.

Yep.

Is I invested in quite a few companies and I was one of the first checks, if not the, or I was in the first round. So I was, you know, one of the first checks, right? If they raise a round of $2, $4, whatever million dollars or $5 million or a million bucks, I was participating in that round. I wouldn't take up the whole round, but I would write a check. And some of these companies started becoming worth $2, $3, $4, $5, $6, $7 billion, whatever over time. And they were growing. But the problem that I ended up facing as an investor is when they start raising rounds like $400, $500 million, you started seeing things like liquidation preferences and all this kind of other stuff that makes it harder for you as an angel to get return versus what I used to see before.

Yeah.

And people were starting to, they started to pay up because when you look at most financial investing, some of these venture funds were starting to get 30, 40, 50% returns. There was a lot of other people who were like, well, we'll pay up for these and we're okay with 20% returns.

Mm-hmm.

And some of the top-tier venture funds do 20% 'cause they have so much money and it's so much capital's being deployed. Entrepreneurs are getting diluted and that's fine, but there's also higher valuations. It's just the angels Don't— from what I've experienced, I'm not saying this is everyone, this is just what I experienced. My returns aren't as great as it used to be 15, 20 years ago. Maybe I'm just picking the wrong companies, but people look at like Anthropic, and I know some of the behind-the-scenes deals and people are like, oh, look at this, they're worth $150 billion. Look at the early angels, they made a killing. No, they're not getting these crazy valuations with, call it liquidation preferences and guarantees on how much return they get, which really does screw up the returns. for the original angels. And I know some of the angels or earlier investors on some of these deals, and I've talked to them about their returns on some of these companies. Like, it's not as good. I'm like, some of these late-stage companies or late-stage rounds, these companies that are putting in, you know, $5, $10, $15 billion, they're not just being like, cool, we'll do it at $100+ billion valuation. Whatever happens, happens. They have terms in there that really protect their downside, which hurts the returns for the earlier angels. So now I just stick with being an LP in the funds. Let those, those guys do it and grind it out. And if I get my 20% returns, great. If it's 30%, great. But the bigger the funds that I'm in, the harder it is for them to produce 30%, 40%. The smaller the funds, like call it sub-$100 million, they have a higher chance. But when some of these funds have like $7 billion, $10 billion just for one fund, it's hard for them to produce 30%, 40%, 50% returns.

They call that the law of big numbers. But here's what I'll say. So, Yes. Has angel investing been like— my very first one was a 50xer, the very first one, right? But here's the thing. A private equity guy recently tweeted this, right? And I was like, oh, this makes a lot of sense. And you and I have talked about this. It's like, dude, if you put money into a fund or whatever, you go into private equity or like, and you sold the business before, you will never have the same returns you had with your business. And that's what we learned. Our business is the best ROI machine by far.

Yes.

I think at least for me, I got distracted by all these things in the past, right? But then you start to look at it. I think a lot of people when they sell their company too, they expect to have the same returns with, you know, by deploying into these funds and all that. It's just not the reality.

So it's not. And like, dude, I've invested like there was one company I put $100 grand in. I think it returned $6 or $8-ish million, $6 and change or $8 and change to me.

Wow.

So it's like I've had some good wins on it and they make up for all my other investments. But the problem is, is like, it's not just picking. I've picked companies that are worth $4, $5, $10-ish billion, and the returns just aren't the same when you're the first check compared to what it used to be. Because now they just raise so much and they stay private forever.

Yep. All right, you can pick one. All right, so I wanted to take a moment to tell you about my podcast co-host Neil's agency called NP Digital, and they work with a whole host of global companies or a global organization. Also, Neil has SEO tools such as Ubersuggest and AnswerThePublic. All you have to do is go to npdigital.com to learn more, and we'll see you on the other side.

Did you see— so there's 2 things and I have— they're both similar topics on here. So there was a post on Reddit. It says, disguised promotion on Reddit for the sake of GEO has to stop. And I saw it on Reddit and funny enough, I also saw that on Tim Soulo's account, the Ahrefs CMO. And I was cracking up when he took a screenshot of it and posted it. And he said, he's like, no, no, no, this is just starting. And I was cracking up because I'm like, this is so true. I'm like, if these LLMs are going to cite Reddit all the time, what the heck do you think marketers are going to do? They're going to spam the crap out of Reddit. And I'm not saying they should. I'm just saying this is the reality.

This is a game for every channel.

Yeah, this is a game for every channel. This is what all marketers do. We've been using spam really loosely, but on Reddit, people really try to spam. It's going to make the community crappier. I'm not saying you should. I think it's shortsighted and I think it's a waste of time of doing that. But Reddit, and I think this was very well due, Reddit stock sinks 12% as ChatGPT references to its content plunges from 10% to 2% in September.

And the headline I had originally, Reddit drops $5.3 billion dollars because of an SEO report.

I believe it was from Yahoo. I saw it from Yahoo Finance. I don't know where it first came from.

So they pulled this. So there's a guy, there's a software called PromptWatch and it shows you the big drop, right? And then some, some, a guy tweeted it and then it got cited by all these news sources.

But no, I think that was great because the problem— I'm not talking about the stock dropping. I don't want anyone to lose money. But when you look at Reddit and the citations that are happening on Google and on ChatGPT, It's insane in a bad way. They're referencing it too much and the quality is really low. I get Google has a deal with Reddit for data, and I get that Sam Altman owns a percentage of Reddit shares, whether it's through his fund or indirectly, however it was. And sure, he benefits from it. And they made an announcement before Reddit went public.

Yeah.

But for me, the quality of Reddit information.

Yeah.

Wasn't—

it wasn't the best to begin with.

No. And they were just citing it like crazy on ChatGPT and it has nothing to do with Reddit. Sure. If I really wanted to, I could figure out ways to game Reddit. I know so many top users, but I, you know, I'm not focusing on that, nor do I care to do that because I didn't believe for the longest time that they're going to continually cite Reddit all the time because of quality. And the same goes with ChatGPT. Dude, you know how many people I know are just guest posting on junk websites and they're doing comparison tables with like structured data, comparing their SaaS product with other people's.

This is no different than the early days of SEO.

Yes. And, and people are like, why aren't you doing that like crazy? It works so well for ChatGPT. I'm like, no, no, no, no. I've seen this show before. It works well right now. Eventually they'll realize this is making ChatGPT data more crap and they'll filter out crappy websites and it won't work the same.

For sure. At best, they will only filter. At worst, they'll start to Ban people.

Yeah, they'll start going through.

Sorry for the—

Yeah, it's okay.

Noise. Neil's drinking a cup of water right now. Yeah. You know what's interesting as you drink? We have a mutual friend here in the Beverly Hills area. So I saw him at a YPO event recently. He's like— and he was kind of talking down to me a little bit. He's like, oh, you got to go on Reddit. Yeah, I'm all in on Reddit. I'm going to go hard on the stock. And I was kind of making the same points that we're making.

Is he the same one who got us both into YPO?

Yeah. Yeah.

Okay.

And then he was like, also like dismissing me as if I don't know anything I'm talking about, right? He's like, no, no, no, trust me, trust me. Sam Altman owns This da da da. I'm like, yeah, man. Like it doesn't help if we spam the internet. We add more crap to it. Like we've seen this with SEO.

Yeah. Sam Altman owns portion of Reddit. Let's talk once he sells all his Reddit shares.

Yeah.

Right? If he already did, I don't know if he's a long-term holder, but eventually he'll look at as oh either I sold all my Reddit shares so I don't care anymore, or B yeah, I still hold shares and my shares are worth fifty million dollars. But man, my ChatGPT experience is becoming worse, and that could potentially cost me. $5, $10 billion in net worth. So yeah, it's all right.

I'll lose the Reddit shares.

Yeah, I'll lose all my Reddit shares or dump them. And then, you know, that's what a logical person does.

Yep. By the way, I mean, kind of related to this one. Oh, so I have this topic here on why SEO attribution is dead. So, but I can't log into my LinkedIn right now. So you can pick another topic.

That's okay. We can just go with SEO attribution is dead and why we agree with that. I don't know what you're referencing there. But I believe SEO attribution is dead for 2 reasons. I don't have the chart in front of me, but when we look at where people learn from a company versus where they buy, SEO is a big chunk of where they learn about a company, but it doesn't pass as the last conversion source a lot of times from where they convert. 2, SEO has evolved into optimizing for, you know, AI overviews, these LLM platforms like ChatGPT, and you're getting the brand mentions. But it's really screwing up your conversion tracking because a lot of people are learning about a company through these LLMs, assuming you can get your brand cited more, but the conversion source isn't always where they first found you. So you got to do, you know, multi-touch attribution. Plus it's not all trackable. When someone sees your brand, they don't necessarily convert right away or even click right away, but they come back. And I think attribution, not just for SEO.

Yeah.

But for SEO, organic social, a lot of these channels are dead and the attribution for paid, what we see is it skews more where the platforms try to take credit for the conversions when I'm not saying they didn't help assist, but they weren't the only touchpoint.

Remember what was said a long time ago with advertising? It's like, you know, 50% of my marketing on the dollar works. I just don't know which 50%. And this is where we're going again, because back in the day when we first started doing this, let's call it like 15, for you, like maybe a little longer, 20 years ago or whatever. You had attribution, you can see keyword attribution, you see this, everything was very much last touch. But you can't look at it that way. And this post actually came from our mutual friend, Gaetano. And like AI traffic could go to your homepage, your pricing, your solutions, like you don't know what's actually driving like, sure, you're good. We're not saying don't look at your analytics. But it's no longer gospel. I think you and I in the very beginning, we looked at it as gospel. It's not like that anymore.

When's the last time you checked your analytics?

I actually checked it like earlier this week. But I was looking at, I was looking at AI traffic data. That's what I was looking at.

I don't think I've logged into my Google Analytics for probably 6 months.

And Neil used to look at it every day. Whenever I went to Neil's house, Google Analytics was pulled up.

I used to look at it like 5, 6 times a day.

Yep.

But on the flip side, I do have a decent-sized team at this point. I don't know how many people are marketing, but we have over 1,000+ people. So there's so many people that log in and send me reports. I don't need to log in. So that's one positive.

To be fair, Neil gets a bunch of text messages in different text groups. That's your team. Yes.

And they break down all the data for me. And some of it comes from Google Analytics. Some of it comes from our own home homegrown solution. Some of it comes from, you know, things like Recurly, Stripe, you get the point. But I do look at data probably 2 to 3 times a day. It's just, I don't have to log into Google Analytics. But I agree with you on attribution. I think there's been a big issue for attribution for a long time. Eric had this headline on his Single Grain homepage for a long time. I'm going to butcher it, but you know what I'm talking— actually, you Better for you to say it.

I think it's 78% of CEOs don't believe that marketers measure ROI. We do.

Yes, it was something like that.

Yeah.

And I think it's totally spot on in which most marketers never really measured attribution and executives never really believed in it. But back in the day, it was easier to measure than it is now. I think it's just, we have more tools to help us, but it's more complex because data's in so many different areas and all the sources don't give you the data. And then the second problem is, is because of that, you don't really know what's fully causing the conversions. So you can show reports, it's just making marketer— it's making a marketer's job much harder in today's world.

Yep.

All right.

So that's it for today, guys. And we'll see you tomorrow.