Meta Just Crossed A Dangerous Line With AI

Eric Siu and Neil Patel discuss Meta's new brain-scan AI called Tribe V2 and its advertising implications. They cover rising infrastructure costs for AI agents, Oracle layoffs, and why Nvidia's profitability matters for the AI industry.
Chapters
Who went up 5% today because they made cuts? I don't know. I didn't see that.
Oracle.
Oracle cut 30,000 people this morning at 6:00 AM. The memo was, it was just, it's like, as of this morning, you're getting this message. We're making a difficult decision to end your role, to terminate your role. You know, we're very great gratitude. We're very gracious for the hard work and dedication that you put in. Please sign a severance agreement in your DocuSign. We'll proceed from there. So 30,000 people cut this morning. Stock jumps 5%. That's Oracle. This is kind of similar to what you're talking about over here.
But yeah, we talked about this around a month ago. Do you know why Oracle made the cuts?
Data center funding.
Yes.
Bingo.
And unlike Microsoft, I know Oracle's a big company, but unlike Microsoft, Apple, Meta, Google, they don't have the same stock cash pile as some of these other big mega tech, you know, multi-trillion dollar companies. So they have no choice but to cut to pay all their debt loads.
You're not like $116 billion in profit like Google where you can just do whatever you want. But, but you know what he can do? He can fund his son's acquisition of Warner Brothers. That's what he can do.
Yeah, but that's more leverage.
I know. Yeah, I'm messing around.
And Google is is not $116 billion anymore.
What is it?
On a quarterly basis, so they're growing at a fast rate. They're accelerating, right? It's $34 billion, so they're roughly at $136 billion a year.
I'm talking about last year's numbers.
Yeah, but you know who's even bigger than them in profit? One of your favorite companies.
This can't be Microsoft.
That's public.
Microsoft?
No, who's your favorite company that's public?
Nvidia. Nvidia.
Nvidia.
Yeah, yeah, yeah. It's my people, you know?
Nvidia right now on a quarterly basis, $42.96 billion in net income. So they're roughly at $171 billion going forward in profit.
It's being stopped, dude. I checked my Anthropic bill this morning. So last week we talked, Friday we talked, today's only Tuesday. Friday we talked about how much did you guess it was?
I guess it was going to be around $6,000, $7,000 as of this morning.
Today's the end of the month. I checked it, $7,500.
All right, cool.
So what's going to happen now? I think we're going to be paying more and more for NVIDIA because here's the thing. I told you last week that I had my agents invade Slack. I'm still holding this mushroom. I'm going to tell you a story in a second, but I had the agents invade Slack. Today, this morning, I'm starting to get messages from people. He's like, hey, I want my own Alfred. I want my own. Everyone wants their own Alfred, right? I'm like, oh my God. I started looking this morning, I was like, okay, I need to buy some Nvidia infrastructure and all this. And the thing is, if I'm gonna decrease my token cost, but you know what? I'm still gonna end up spending way more anyway 'cause everyone wants it, right? It's all your, it's, I wanna say they're your children, but it's all your people that you wanna help. It's your tribe that you wanna help, and everyone should get a token budget. So it's becoming, I'm like, oh my God. And you're seeing the same thing too.
Well, when you look at Nvidia specifically, 'cause Claude Code and all this kind of stuff that's came out, OpenCLAW, I think Nvidia's version is NeoCLAW or something like that.
NemoCLAW. NemoCLAW.
Like the fish. Okay, cool. So in January 2025, if you look at year over year with publicly traded companies, they look at quarter over quarter. So Jan 2025, they would compare Jan 2024 earnings. So they're at 77.94% growth. Of course, at their large numbers, growth is going to slow down. Then April 2025, growth slowed down to 69%. In July, 55%. And then what happens? Claude, with all this stuff, starts pushing October 2025. 62%. January 2026, 73%. So Nvidia's growth is now reaccelerating because all the changes. This just tells you in business and marketing, when we're talking about AI, it's an easy way to see if there's major catalysts. And what we're seeing with agents is starting to become reality because we've all been talking about agents for a year or 2 years, but a lot of it's been—
excuse my language. Pie in the sky.
Yeah, pie in the sky. I was going to say bull. And now It's really becoming a reality where you're seeing people, not just companies, use agents. A lot of people can afford it because it's really cheap and it's doing stuff to make them more efficient, better, and grow faster.
Yep. So I'm going to— we're going to come back to AI. There's, there's plenty of AI to talk about. Look, by the way, look at all the topics. You see the topics that don't have my name next to them? Those are all OpenCLAW generated, by the way. I was like, let's do some classic Marketing School episodes. I'm like, give us some things that we can clip, right? But anyway, I'm hold— I've been holding this chocolate mushroom tip that Neil's given to me. It's called— what's it called, Neil?
Cocoa Mound or something? I don't know.
Cocoa Rooms or something. Choco Rooms. Choco Rooms. Yeah.
Okay. Why is it a chocolate mushroom tip? It's just a mushroom.
Now we'll leave it at that. We'll leave it. We'll leave it for other people's imagination.
Quick break.
If you want to run personalized LinkedIn ads and have personalized landing pages to convert your customers at a much higher rate on LinkedIn, check out Karrot. That's K-A-R-R-O-T dot A-I. Karrot allows you to do things that LinkedIn Ads does not allow you to do right now. Check it out. There are publicly traded companies such as Semrush and Sitecore that are using this right now.
And you can use this to get ahead because if you don't use this, it will take you weeks, if not even a little longer than that, to make these ads.
Because again, you cannot do this on LinkedIn. So again, go to www.karrot.ai and we'll see you on the other side. I don't know if I've been forwarding you this stuff, but you see from a lot of doctors and they're like, oh, you know, so, so-and-so microplastics, like The toilet paper that you use to wipe your butt, microplastics, forever PFAS, forever chemicals. And what the Bounty tissue that you use to wipe your hands, like the paper towels, right?
Yeah.
All PFAS, right? And they're getting tons and tons of views. Like, I see everyone in the comments are like, I guess I can't buy that anymore. And you know what happens immediately? I go to Amazon, whatever they recommend, I go to Amazon right after and I just order a bunch of new like bamboo toilet paper and things like that. So, I'm just like, negative marketing does work. And these scientists that I don't think they're doing this on purpose, but they're helping all these other companies grow. And so, what does it have to do with this chocolate mushroom tip over here?
I'm curious.
So, my point is you talk about anything like this. Oh, it's like the packaging, microplastics, it's dangerous for you, right? You can't eat it anymore or dairy, you can't have it anymore. I'm like, okay, I don't want to buy it anymore, right? It's like negative marketing and this is aside. I asked Neela to bring this down because this tastes really good but we get so much marketing around what we can't have anymore. I think at a certain point, it's just like whatever. If you can do the super organic alternative, you end up doing very well, which is why the toilet paper brand I ended up buying, Real. So I bought this new toilet paper and I bought these new paper towels. Amazon Basics now has bamboo toilet paper too. So number one, negative marketing works. But number two, sometimes you don't give a damn because it tastes so good. So that's right. Yep.
And the way I look at marketing, one of my kids was telling me the other day that we were at a gas station, we were driving back from Las Vegas to Los Angeles. And they wanted some chocolate. And I was like, you can pick one thing from the gas station. Just don't make it too bad, right? Like try to be somewhat health conscious.
How do they know what that means?
They know, they know, like what's healthy, what's not like carrots and broccoli or hummus. And, you know, they know eating just a big bar of chocolate candy is really bad, but eating a smaller bar of chocolate is still bad, but not as bad because it's just, you know, not as much fats and all that kind of stuff. And My daughter turns to me and they say, well, if we're going to eat something bad, might as well eat something that's really tasty and bad. Because, you know, if we're going to eat something that's unhealthy, I'd rather enjoy it the most. And I'm already going to eat a dinner with vegetables. And she's telling me this. So she's just like, I want what I want. Is that okay? And I said, sure. It just can't be like a really big size, like, you know, like the king size and stuff like that. And when you think about it from a marketing angle, Some things are bad for you. It's okay. And a prime example of this is chocolate candy. If you're going to eat chocolate candy, you like it, might as well eat the good stuff. Like, why would you eat the Kraft chocolate candy? Like, what? It's not as bad for you? Who cares? Just eat it.
That's why when I go bad, it's called— my friend taught me this. She calls them fatter days. So on Saturdays, I do fatter day. And I'll get like not one double burger trip, but I get a triple burger, Noah, and I add pastrami to it. And then I get another one on top.
Seriously?
Yeah, that's like 1,500 calories, but it's okay.
So, you do it every Saturday?
Either. So, I either do that for Friday or I do the Erewhon steak breakfast burrito. And then, I also get like the chocolate toffee pistachios. So, I get 2 cans of those.
Oh, that's cool.
Yeah.
I tend to—
What do you do?
None of that. I tend to just be a garbage disposal and eat whatever my kids have leftover.
Oh, okay. So, you have junk food every day then, like a little bit?
Not much every day. Like if I had to put it in a calorie perspective of how much food that I'm eating that's junk Maybe like 50 to 100 calories max.
Yeah.
So, it's not much. But the way I look at it, it's like when I came down here, you're just like, dude, no mushrooms? I was like, what do you want? So, I just went upstairs and I got you more than the mushrooms because my kids have so many, you know, like snack food items in the house. It's like, cool, I get you whatever you want.
You don't do like the fruit snacks anymore anyway.
So, I do. Those are healthier versions.
Got it.
And then we also have fruit snacks, but next time I'll have to bring fruit snacks.
They tend to be no food color.
So we don't do like, you know, food coloring or any of that crappy stuff. They are unhealthy for you, but you can read the ingredients and understand what they are.
So we'll go back to marketing. So Neil, let me ask you this. What do you think the highest paying customer acquisition job is today?
Highest paying customer acquisition job in marketing? So I'll let you give— I'm assuming you're not talking about like a CMO.
You're talking about like specifically like an individual contributor.
I would either go someone who's running paid or someone who is running SEO.
Okay. So interestingly enough, it's neither of those. So this is where we can react to what Kevin O'Leary said. So Kevin O'Leary thinks that the highest paying job right now is customer acquisition on social media. So he's like, I used to pay these guys $48,000 a year. Now I'm paying them $250 grand a year because you can measure their work based on customer acquisition every week. Most of them are contractors. They make half a million dollars a year because they know how to make content, turn it into a 59-second ad on social and acquire 200 customers. These people in their early 20s are so valuable right now. If you know how to use your phone, somebody wants to hire you. So I would say caveat here is it's organic social, but I think it converts. They also will convert to paid as well and handle that, that side of the funnel.
But obviously, you know, Kevin is more consumer focused and all that personal brand and he's looking at it from an affiliate model in which I'm assuming his compensation is based on the results they drive based on how it sounds because he used to pay $40,000.
Knowing how he does deals, probably. Yeah.
Yeah, I can see him being more performance based. In almost all companies we work with, organic social is not the highest paid person from an individual contributor standpoint. Typically for large corporations, the budget's going to Google and Meta. If you don't believe me, just look at their stock prices. You can see how they're getting the lion's share of the marketing dollars and people tend to get paid the most to manage where the money's going.
Yeah. So I tend to agree with that. I mean, also we're biased because that's what we see with our businesses, but It's not, no, it's not just with our businesses.
You see it with your clients too. And I see it with my clients.
I know, but it's the data that we have and we get, but—
Yes.
It's what's in front of us, right? So—
It's what's in front of us. But I would say we see most marketing data, like we see a lot of offline data. You see offline data. It doesn't matter if you and I don't offer all the services. We work with so many enterprises and just corporations. We see data in different sectors and in different fields and what companies are spending to a lot of different agencies. But the highest paid role is typically the people managing the money, if you're not talking about C-level roles like a CMO.
So you just mentioned Meta, by the way. So Meta just dropped something. I don't know if you heard about this, but they dropped something bigger and supposedly more dangerous than what OpenAI has ever built. So Meta dropped an AI called Tribe V2. It's basically a digital copy of your brain. So here's what it does, Neil. So they trained it on 1,000 hours of brain scans from 700 people lying inside MRI machines. It doesn't read your thoughts. It does something worse. It knows what's going to make you feel something before you even feel it. Think about that for a second. If an AI already knows which image, which sound, which word is going to hit your dopamine, you don't need to read someone's mind. You just build the perfect trap. So the whole idea with this is that this— they don't have to guess what will keep you glued to a screen anymore. They can just rehearse it and they already have all the data, right? So they can just make whatever they want now. And they'll probably make this available to the advertisers too, to get their advertisers to continue to spend more and more. Yeah.
And dude, have you been getting hit up by— I always mess up their name— Manus or Mantis? Manus.
Yeah.
Manus. Is it Manus or Mantis? Manus. Manus. I think there's a biotech company called Mantis Biotech or something like that. AI, right?
Probably right.
But either way, I've been getting hit up a lot by their team and they're just like, hey, we're looking to pay creators to push Manus. Facebook bought us and we can do a lot of the stuff OpenCLAW does. You don't need to be a developer. We'd love for you to create content on this kind of stuff. It's just interesting to see how Meta has so much reach and they're still going after influencer marketing for this stuff. And the first thing that when I was talking to them, I was like, why don't you guys just run ads on your own platform? It'll be easier and you'll meet— hit a lot more people than if you're just going and leveraging this influencer stuff. And I honestly don't think the influencer stuff is that effective because you have to label it as paid. The moment you put that disclosure in there in the beginning, dude, every time we look at stats from it, the drop-off is just ridiculous. Now, it doesn't mean influencer marketing isn't profitable and it can't work. It just doesn't scale to the way that paid— no, yeah, that paid works. And even like organic SEO or GEO works. But it's more so, remember that startup, was it Cluey? It was a Korean founder. I think it was from like Stanford or something like that. Bright kid.
Top dog, yeah.
Ended up creating Clue.ly. And I think there were some articles that the revenue numbers, he eventually admitted that his revenue numbers weren't really the revenue numbers.
Dude, he's like full-on influencer now on Instagram. Do you see his stuff?
I haven't seen his stuff. I don't follow him. But he used to break down like influencer marketing, where it's look how much we're growing. And he was getting everyone to start mentioning him, but yet he wasn't really generating that much revenue from it. But on the flip side, I have a buddy who owns a company called Legion Athletics, the protein supplement company. They do extremely well with influencers. Their model's quite a bit different. They look for people who are diehard, you know, into lifting and sports and, you know, like organic, clean products. And they pay them. They may not have the biggest following. They don't get paid as much because of their following size. That model works extremely well for them and it scales and it's profitable, but it's good product targeted with— The right influencers, not just influencers with followers. But yet when everyone wants to do influencer marketing, they just care about follower count, view count versus finding the right audience.
Dude, my agent now, when people reach out to me for the influencer stuff, they'll be like, oh, the engagement isn't what we're looking for. We want to pay on CPM. It's like Mr. Siu will not accept anything lower than $25,000. It just keeps responding like that. And then it keeps asking for different angles. And then Mr. Siu is like, what is your new proposal for Mr. Siu? It has to be higher than $25,000. So it keeps saying like different angles.
Does it update you with all the messaging when it's negotiating?
Yeah, it's, it's, it'll like give me an end of day update saying, here's what I, here's what I'm sending back, what do you think? And here's what needs your approval as well. I'm like, this is good. Would you like this?
So I was talking with Sam from WriteSonic. He was showing me his version of it when he deals with influencers. He then gets a Telegram message and then he can end up, it'll say, we are going to counter with this. Is this approved? And he goes, yes or no. And he creates boundaries. He uses OpenClaw. And he's like, do you want this? And I'm like, no, I'm not interested in doing influencer deals. Even with Facebook one, when they hit me up, They're like, cool, would you be interested?
Yeah.
And I was like, I'm like, one, what are you going to pay? And then they're like, we'll get back to you. And they said, what are you looking for? And that was the 2. And I said, I actually don't care to be paid. How can you drive me more business? That's more important to me than you pay me. Because it's true. And this is why I don't respond to the influencer deals. I don't care for the income. I want more business.
Why don't you just have an agent that responds like that? Hey, what are you offering? Oh, actually, I don't care for that. Just send me the leads.
But it's not that simple because most of the people that hit us up for the influencer stuff, let's say it's TikTok, TikTok is doing an influencer campaign. I'm making it up because we do get hit up, but it's never TikTok reaching out to us. It is an agency on TikTok's behalf. And Meta was a little bit different because actually the Manus team hitting me up and they are part of Meta.
So, okay, this is the annoying part, but I remember I just did a deal with LinkedIn recently, right? So it was an agency that reached out, but then I got on a call and LinkedIn was there. So then I added both of those people and then I have the relationship. And then you can play like that, but you still need the human aspect. You can't just like offload it right now.
And it's hard to go through directly to the LinkedIns, not because they're LinkedIn, but you have a different issue. A LinkedIn, for them to get you in their system and then to get you paid and booked as a vendor, it's a pain in the ass. So they go through agencies. So cutting out that middleman is going to be really tough, but they usually are like single track. Like, hey, just go find us influencers. It's hard to actually do a mutual.
Yeah, yeah, no, that's right. That is it, guys. Don't forget to rate, review, and subscribe, and we'll see you tomorrow.










