00:00
Alright, everyone. We just finished recording this pod, and we wanted to come back and give a little bit of an intro because
00:05
I feel so many different emotions. And here's why the stakes of this are super, super, super high. Yeah. We're talking about what's going on with robots in the real world.
00:15
So robots that are being used, we have specific examples of really cool businesses that are doing robots for,
00:21
you know, lawn care, restaurants, all different types of things. But then we go over the Elon Musk versus Sam Alman lawsuit. What is this lawsuit between two of the power players of Silicon Valley and what's our take on it, which might surprise you, as well as some of the brilliant the nuance, the the marketing and PR that's going on here, we kinda call it out. This is actually we have fun while doing it, but this is actually really, really in my opinion. Listen to the entire thing, and you have to listen to the first half to build up the story and the stakes for the second half. Trust me. You're gonna understand exactly what I'm talking about when you get through the whole thing.
00:53
So enjoy it.
01:02
Alright. What up, Sam? Hey, what's going on?
01:05
Sam, I gotta tell you something.
01:07
When it comes to kissing,
01:11
eating ice cream before bed
01:13
or business.
01:14
Timing is everything, my friend. And I think I think that timing
01:19
is important in all aspects of business. And in technology, there's a moment happening right now. I could feel it. I could feel it. My like when you hit your funny bone and your whole arm gets hot, I could feel it right now. Shits about to change.
01:32
Robots are going wild. I gotta talk to you about robots. And they've, like, been around for a minute, but now it actually feels like it's a reality.
01:39
Exactly. So right now, a lot of people are talking about
01:43
Tesla because Elon Musk came out, and he said,
01:46
the future of Tesla is this humanoid robot.
01:49
And I forgot what they call him. Like, optimists or some shit like that. So it's like, he's like, hey, we got this robot that can walk around, pick up stuff, do tasks. It's gonna change the workforce.
01:59
And,
02:00
your boy, Brad Adcock. He's got a company called Figurate, which you're an investor in. They have their robot. I think they just raised, like, a billion dollars or something. Right? Like, that's last week. I think they raised six hundred million dollars at a two and a half or three billion dollar valuation, I think. Nice little investment there for you.
02:16
I know. I was in the first I was in on the first round. I doubled down and did on this round. Is this gonna be my my Motorola? Is this gonna be my my big win? Could have been anybody's because you had him come on the podcast and tell the world what he was doing. So,
02:30
good good on you. You multi billion dollars,
02:33
valuation right now. So that's a big deal. And people are really excited about that. I think there's these stats that say
02:39
if, basically, labor is fifty percent of the GDP. The labor market is a ten times bigger market than transportation, which makes sense why Elon is like, hey, forget cars. It's all about these robots. We're just we're an AI company. We're doing robots.
02:53
He's very good at repositioning
02:55
and re
02:56
imagining the vision of the company over and over and over again. Right? And went from affordable electric cars
03:03
to self driving autonomous vehicles, robot taxis that will just drive people everywhere. Your car will become
03:08
an appreciating asset instead of a depreciating asset. And then now it's actually it's not even about cars. It's about AI robots, and that's that's the future. But I wanna tell you about something that's less fancy than that.
03:19
Because I think it's gonna come first. And that is single purpose robots,
03:24
aka,
03:25
roomba's on steroids.
03:27
And roomba's on steroids, you know, the roomba was a it's a single purpose vac little vacuum robot that just goes around your house. And it but it's in the wild today. It, like, actually works actually useful for people. An even better example, I think, is the dishwasher, which is in pretty much every home in the United States right now. And it saves you a ton of time. It like, when you're, like, my dishwasher broke last week, terrible experience when your dishwasher breaks. The dishwasher is a single purpose robot. That's actually out there in the wild. We should be striving to create something as useful as the dishwasher. That should be the north star right now for a lot of lot of companies. And there are. So I I went on Twitter today,
04:04
and I saw this great video
04:06
that was about,
04:07
a company called Electric Sheep. Have you seen this?
04:11
No.
04:12
Electric sheep. That's electric sheep. I I don't know what it is. That's a great name, I think. And I think it's a lot a lawnmower.
04:19
It is imagine if you took the Ruba
04:22
and you put it in the lawn. That's what it is. It's a That's a great name. It is a robot that mows your lawn. Does all kind of land scaping. What's interesting about these guys that I found pretty fascinating is not only are they a company that's, like, a bunch of guys in a garage building a robot, like, a lawn mower that's got camera on it that can go around your lawn, they're actually buying up and rolling up landscaping services. So it's a combination of a sweaty startup. And a moonshot startup together. So these guys are going out. They're buying all these landscaping businesses in the Bay Area. And because they're like, well, we need a, like, a testing ground full robots. Like, you have to actually, like, in their office, it's just like an astroturf. They're like, yeah. Well, it can do this square astroturf pretty easily. It's not so bad.
05:03
But in the real world, your lawn has got all kinds of different lumps and dips and divots and trees and rocks, and it has to be able to work. How are we gonna get the best data? Right? Just the same way that Tesla is putting all, like, the cars on the road, collecting data at all times, and they're collecting, like, whatever. A hundred times more data than the next biggest self driving car company. These guys are trying to do that with their lawn care robots. So they're buying these landscaping businesses.
05:25
Then they use them for as a sandbox, they take their robots out, and they try to use them they mow your lawn normally, but they'll try to test their robots as well. They gather the data. And then over time, they're like, well,
05:36
we're buying these landscaping
05:37
businesses at one valuation.
05:39
But then what we'll do is we'll st slowly start to replace the workers in the landscaping business with the robots and actually have improved margins.
05:47
And so we're gonna, like, it's like a private equity play as well. And so I thought that was a pretty fascinating approach that I hadn't heard before. I can't decide if that's just overkill, like, Maybe that's not actually necessary to to do to do two hard things at once, but I did find it very interesting. What do you think of that strategy? I'm looking at their website. Alright. So It's slick. This is cool.
06:08
I agree with you. The whole, overkill thing that bothers me. That could be true. Well, the problem is, like, The the founders, like, founder, like, went to Stanford, and, like, you're smart enough to build robotics company. That's almost you're smart enough where you can get yourself into trouble.
06:24
Right. You can you could be like, oh, we'll vertically integrate. And it's like and then use the word words vertically integrate?
06:30
Be real careful.
06:31
You might be midwitting the shit out of your plan right now. Basically, like, for him, there needs to be a Venn diagram of this guy, this founder of, sheep robotics, he's, like,
06:42
the dumbest smart person, and he needs to hang out with the smartest dumb person.
06:47
And we need that, like, We need that, like, event diagram overlap to to make this the perfect business. That's that's kinda like what he's doing.
06:54
This is awesome. This is awesome. This is cool.
06:57
Another thing that I worry about.
06:59
This is badass.
07:01
What is it a badass
07:04
team and a badass execution?
07:06
On, like, an small opportunity.
07:09
Yeah. Like, lawn I mean, I get I don't know. Is lawn mowing a bid big I mean, is it as grand of a plan as what these people are actually capable of doing?
07:18
I think so. Because I, you know, I don't know how big the landscaping industry is, but I know that Anytime private equity is going on rolling up things, there's a lot of money to be made. They they wouldn't private equity doesn't roll up, you know, like,
07:29
lemon you know, small lemonade stands, for example. Like, they go into things that are that that are cash flowing, you know, industries. And so I do think if you added up all the landscape businesses, and you're like, well,
07:40
you know, either what they're gonna do is they're gonna use them in their own business to get it right. And then they're gonna go lease or sell these robots to every landscaper in the country and be like, hey. Would like to improve your your operating margins by, you know, ten percent, twenty percent,
07:53
use these robots. And so that either that's what they're gonna do or they're gonna go try to compete them somehow on price. I don't really love that idea. So so I think it has to be. They're just using this to prove their own tech to then go lease these out to every other landscaper.
08:05
So on their about page, they list their team. And I guess,
08:09
their advisors include this guy named Dan Foley. It says Dan Foley was a CEO of a company that sold to Brightview.
08:16
I googled Brightview, Brightview is a commercial landscaping company,
08:20
that did roughly
08:23
three billion dollars last year in revenue. Yeah. Publicly it's publicly traded. It's it has a shit market cap. So what you're saying is the newsletter guy
08:31
is calling
08:32
the land guy small?
08:35
No. What I'm saying is I yeah. I did call it. What's a boy? What's a boy?
08:41
I did call it small. And then I looked at Brightview, and it just shows I'm I I'm I'm I'm an idiot. I don't know anything. Like every great man, you're willing to change your mind.
08:49
Congratulations.
08:51
I I'm only one Google away from saying sorry. Alright, everyone. A quick break to tell you about HubSpot. And this one really easy for me to talk about because I'm gonna show you a real life example. So I've got this company called Hampton. Join Hampton dot com. It's a community for founders doing two million all the way up to, like, two hundred fifty million dollars a year in revenue. And one of the ways that we've grown is we've created these cool surveys. And so we have a lot of founders who have high net worth, and we'll ask them all types questions that people typically are embarrassed to ask, but provide a lot of value. So things like how much the founders pay themselves each month, how much money they're spending each month, what their payroll looks like, If they're optimistic about the next year and their business, all these questions that people are afraid to ask, but, well, we ask them anyway. And they tell us in this anonymous survey, And so what we do is we created a landing page using HubSpot's landing page tool. And it basically has a landing page that says, here's all the questions we asked. Give us your email if you wanna access it. And then I shared this page on Twitter, and we were able to get thousands of people who gave us their email and told us they want this survey. And I could see, did they come from social media. I can see did it come from Twitter from LinkedIn. It's basically everywhere else that they could possibly come from. I'm able to track all of that. And then I'm able to see over the next handful of weeks how many of those people actually signed up and became a member of Hampton. In other words, I can see how much revenue came from this survey, how much revenue came from each traffic source, things like that. But the best part is I can see how much revenue came from it. And a lot of times, it takes a ton of work to make that happen. HubSpot made that super, super easy. If you're interested in doing this, you could check it out hubspot dot com, the links in the description, and I'll also put the link to the survey that I did. So you can actually see the landing page and how it works like that. I'm just gonna do that call to action then.
10:24
And it's free. Check it out in the description. Alright. Now back to MFM. So check this out. Okay. So that's the first one. Now I have a similar, you know, what where else are robots being used? Again, these single purpose robots, not the general robots, but single purpose robots. In the same way that we got specific AI before we're getting Artive AGI artificial general intelligence. We got specific AI where it's like, you know, I don't know. It's been, like, twenty years since, deep mind could just beat Gary Casperov at chess. Right? Like, we we solve chess, then we solve Go, then we solve math, then we solve route planning. You could do these single purpose AIs that way before you're gonna get to general AI. So similarly, I think the timing of the opportunity is not in the the tesla robot or even, sorry, the figure robot, which is, like, just a human robot that could do everything.
11:11
You know, that is the big biggest win, obviously. But I think before then, almost by definition, we have to see these specific single job one job robots. Because you get it's gonna be easier to build a one job robot than it is gonna be to build a general robot that could do all jobs. And so, here's some other ones.
11:28
So I don't know if you know this, Sam, but This podcast has not only listened to by people who are trying to make their first million, but maybe by people who are trying to make their their hundredth million. They're they're neck their next million.
11:39
There's more millions to go. And one of those people is the CEO of a of a restaurant called,
11:44
sweetgreen.
11:45
And so the the CEO of sweetgreen is a listener of the pod. And I was looking where else are robots being used? And I saw that suite green actually now has a, what they call, their infinite kitchen, which is a, like, robot driven version of sweetgreen. So sweetgreen is like a salad,
12:00
bowl type of type of concept. You normally go in there and there's, like, you know, whatever four people working the line, and then there's, like, four people in the back kitchen doing prep, just like most fast casual restaurants. And what they've done is they basically went in and they were like, hey. What if we this first, I think they bought a robotics company for fifty million bucks. And then they use that technology to build of concept where you basically have, like, one person in the store and the rest of it is all done by robots. And so click this link that
12:28
I put in our sheet. We could show this on YouTube.
12:30
By the way, the is this Jonathan Newman?
12:33
Yeah.
12:34
I know that he's a listener. You wanna know how I know he's a listener? Tell me you're getting free sweetgreen, and I'm not. No. Even better. He emailed me, and he took a screenshot and he showed that he subscribed
12:44
to my first million on both
12:46
YouTube
12:47
podcast for Apple and Spotify. So if you wanna be like John and start a multi billion dollar company, you gotta subscribe. That's all you have to do. I don't think there's been anyone who subscribed that isn't rich and famous and successful and desired by many people of the opposite sex. We're the same sex if that's what they're into. It's just a thing that works. Alright?
13:06
Okay. So so sweetgreen, so check this this video out. So, basically, what they did was they made these, like, this kitchen. It's just like a giant wall where all the ingredients are in these, like, little robotic kind of dispenser arm type things. And whatever you order, the bowl goes across the conveyor belt. And then it's, like, the bowl is spinning. It's, like, rotating so that it drops in all the ingredients at the right spots. And, again, they have all kinds of ingredients. Right? So, like, this sounds pretty simple. It's just, like, dispensing. But, like, maybe dispensing almonds is really easy, but feta cheese, which is really crumbly. You gotta be careful with it. Right? Like, so you said, they basically fine tune this thing. And now this concept can do five hundred bowls per hour.
13:44
I deformed him. And I was like, how much does, like, the, like, a human, like, your normal restaurant line do? He's like, maybe, like, one seventy to two hundred. Wow. So you get basically, like, more than two x the speed.
13:56
And now
13:57
you have a robot that does work that can work twenty four seven. Never calls in sick. Never, you know, always get puts the right proportion into into every bowl. You know, you have perfect food control cost. And so you end up with a better version of an employee. And the strategy, the way that I think about this is you trade OpEx for CapEx.
14:15
And trading OpEx for CapEx is actually a pretty good idea.
14:19
I think you should further explain what that means. So today, all restaurants work with human labor. Your human labor is your operating expense. It's the OpEx. And so and what that means is that for every
14:29
hour that you're operating, every every bowl that you're selling, you need to pay a certain amount in wages to to employees. And so let's say today, you know, you're paying twenty bucks an hour to people that work in these restaurants, you got you know, ten people working in the restaurant at any given time, blah, blah, blah, you can add it up. And let's just pretend that your labor costs come out to thirty three percent of your revenue.
14:50
And so the opportunity here, which is, I would say, normal for for a restaurant, food costs usually thirty percent, labor costs usually thirty percent.
14:56
And so you you you add these up, you're right. You can't change the food cost, but could you change the labor cost? And so what the what he did with these restaurants is every restaurant to put in these robots cost more money. It cost more CapEx, which is, like, the initial upfront investment. And I think in their last earnings call, they said it's about five hundred k to
15:15
put the robot in. So you put an extra five hundred k to build the store. But again, this is five hundred k of CapEx, which means you can depreciate it, which is good for your, you know, tax purposes and whatnot.
15:25
And it's a one time cost versus your OpEx is a everyday of of operation. And what they found was for that extra five hundred k investment, They get at least an extra seven percent of, you know, of net margin from, from those restaurants that because they have less optics. And so you would take that It that basically creates, you know, some breakeven point where you say, okay. Cool. Two years in, we're gonna have paid off, you know, the cost of this robot or whatever.
15:49
And also, he's like, you know, the five hundred k CapEx, that's today. And this is the most expensive it's ever gonna be. Like, the costs are only gonna drop as we scale up as we refine the as we simplify the design, like, every version we come out with is just gonna be lower cost than this one. And I asked them, I go, in ten years,
16:07
What are the odds that,
16:09
your restaurants are basically robot robot driven not human labor? Is this guy sweetgreen's publicly traded. Right? Yes. Is he are you allowed to say what he's gonna say right now? I told him I said, is there anything in that I can't share in this? And he goes, no. I've shared all this. But it was basically, like, it's very high. Right? Like, and just just thinking about that a little bit out loud. It's like, man, that means that the next ten years in my, you know, very easily,
16:31
in our lifetime.
16:33
Restaurants are just gonna go for you know, a lot of restaurants are gonna go from. You walk in and there's just, like, a bunch of people who are
16:39
Sort of disgruntled people who would rather just be checking their phone right now than, you know, sitting there with this, like, stupid uniform on. And it's just gonna go where you go in, you push a button And then your bowl gets made by a robotic arm, and then you walk out. Like, that's gonna be things. There's gonna be one human in the loop. One one human whose job is now called you know, robot maintenance, robot operations, and they're basically the guy's like, oh, the carrot's machine got a jam. Let me go and, take a little fork and unjammed that thing or Hey, I gotta refill the container of apples. And so what the what these guys do with this week in restaurant is the humans are all in the back doing the fresh prep of ingredients. And then the robots are serving the customers. And that's how they split it today. So you and I are are fairly optimistic.
17:20
You know, we try to see the the bright side of most things. You in particular do that, but let's try to be pessimistic right now a little bit. So the argument that you're gonna see from a lot of actually, most people, I would think, but particularly pessimists, but most people are gonna be pessimistic on this. They're gonna say, But this is gonna ruin the economy. This is gonna ruin jobs. And let me ask you a question before you actually give your opinion. You have
17:45
Sean, you own a business that employs warehouse workers.
17:49
Roughly, what's your annual turnover for your warehouse? Workers. Like, how often do they go? We use a three p l now, but before when we when we had it in house, it was
17:59
I would say every year,
18:01
we're gonna replace ninety percent of the team.
18:04
So I was reading I think it was Amazon. I forget someone who employs ten or hundreds of thousands of warehouse workers. For them, it was It was one hundred percent.
18:13
So it was
18:14
one hundred percent turnover rate. And I remember talking to Brett, from Figger, and I go, Brett, Are you gonna, like, are you preparing to, like, be the most hated man in America right now? And he was like, well, why? And I said, well, because, like, you're gonna be re you're get you're taking jobs. And he was like, well, I'm actually taking jobs. In fact, there's all these warehouses that no one will no one's gonna work there. They just literally cannot cannot convince people to work there whether they can't afford it or if it's just a people,
18:42
aren't interested in that type of work. And then and then he goes plus the average turnover on a lot of warehouses, it is a hundred percent. So, like, they just can't keep people, like, in these jobs for a long time. And so my question to you is, are you pessimistic at all of what's gonna happen in fifteen years, twenty years? You know, my first job, I was a laborer and then a dishwasher.
19:02
I bet I think you probably have done similar things like that. What's gonna happen? My first my first business was a restaurant business. I was the guy chopping tuna and doing all that stuff in the back, doing dishes, you know, every every night,
19:13
I think there's two versions of pessimism. The first version is it's not gonna work.
19:17
The second version is it's gonna work, and that's a bad thing. You're talking about the second version. I think this I think I think it's gonna work. It one hundred percent is going to work, I think. But I I asked the guy. I was like,
19:29
How I was like, what's the non obvious, like, only when you're in the weeds doing it? Is it really obvious about this? But to the outside person, they maybe gloss over it. I was like, what's what's that in That's a great question. He goes, well, it's like it's like most things in technology because the first ninety percent is easy, which makes it really easy to have, like, a sweet demo
19:48
And everybody just immediately is like, oh, man. This is gonna change everything.
19:51
And it's like, it will eventually,
19:54
and probably a lot longer than we we we want.
19:58
Because he goes, the so the ninety percent is easy.
20:01
The next nine percent is ten times harder. So it's like, you know, the simple stupid example of this would be like, you know, dispensing almonds, easy, dispensing the right amount of crumbly feta cheese,
20:13
it is literally ten times harder than the all men's problem. Right? Or, like, on a Tesla, like, my Tesla self drives, and it's the autopilot is mostly good unless it's a flashing yellow light, then it, like, freaks out and I almost die. Right. Or it's, like, if it's a dewy morning or it's a foggy, this, whatever, or it's rainy here, or this instruction there. Right? There's, like, all these edge cases. And then that's what he said. He's, like, the last one percent. Is this even ten times harder than that first one? And he's, like, so with self driving cars, the problem is Nobody wants a ninety nine percent good self driving car, even if that's actually five times better than a human. Yeah. It's like if I'm gonna die, I'm gonna die on my own hands, not at the hands of this, you know, this piece of software. And so that makes it really hard to roll out these things and gets and and deal with the PR backlash. Every time there is a a
20:59
crash or a or a a a bad result,
21:02
even if it's less pro even if it's a better probability than than humans driving. But I don't think you have the same problem with restaurants. Right? Size of the price is smaller, but also the barrier gentry is the barrier to deployment as much smaller, where it's like, oh, man. I asked for light sauce. This is medium sauce. Right? Like, you know, it's like, whatever. It's okay. It's not, self driving car crash. And so I do think that that's the case here where
21:24
a lot of people can get ninety percent of the way they're getting the next nine percent ten times harder, getting the final
21:29
one ten times harder than even that. In software development, we used to say this where I would always be because I'm the non engineer. I worked at a company of all programmers, and I'd just be, like, walking around, you know,
21:40
eating cashews and just walking around just like, hey, hey, guys. How's it coming? When when do you think it's gonna be done? And they're like, well, we're about ninety percent of the way there. We got, like, ninety percent of the things done. So we're about halfway done. You're like, great. Ten minutes will be done. I was like, wait. Ninety percent done. So you're halfway done. So, yeah, the last ten percent always takes at least double the time that you've already spent. I was like, alright.
21:59
Cool.
22:00
That's how that's how math works.
22:03
I, if you remember coffee, I'll go get some. Like ten years ago, universal,
22:08
income or, you know, universal wages was like a thing. UBI.
22:12
UBI. We're gonna you're gonna we're gonna give everyone in America a a bunch of money And, like, I'm a capitalistic guy. And I was like, what? If that, that's the worst. And then I learned two things. One, do you know, in last in Alaska, I've had a friend I've had friends move to Alaska. You know, they do that. They've been doing that for decades. They give you five grand. The oil oil dividend. Right? It's like you get a share of the money that they were making from having so much natural resources, I think. I don't. Yeah. Maybe I don't know where it's from, and I I believe if I'm not mistaken, I think sometimes it's a tax credit where, like, you just pay less in taxes or if you don't owe an any tax money, you literally get a check of, like, forty eight hundred dollars or fifty five hundred dollars, something like that. And then the second thing that I learned was I started researching UBI, and I was like, wait, why is Sam Altman advocating for u UBI? Like, why are all these, like, really smart robot makers advocating for it. And I'm like,
23:03
shit.
23:04
They know.
23:05
They It's like When my wife comes home and my Why is the kitchen clean? What did you do?
23:11
Why are you cleaning up this mess? You did something else. That's what he's doing by the starting UBI.
23:16
Which he has. He has. You know, I think Sam Altman along with a few other guys are doing experiments in San Francisco where I think they're giving eight hundred dollar. I forget I don't even wanna say because I'm gonna get the numbers it was like a a a fairly small stop. They stopped now. They I think they did it for, like, some number of years. And that was the experiment, and then they stopped. And I've we should go look up actually what actually happened. But I remember reading something that was, like,
23:38
not, like, entirely positive. It wasn't, like, oh my god. This is great. We this is the program we need to roll out nationwide. It was, like, Well,
23:47
no. That's not that's not what the takeaway was. But it's still an interesting concept to me that I'm open to learning more about because when I learn more about these robots, I actually do think I'm like, I'm just old enough where I think this isn't gonna impact me. But for most Americans who are born, who are young now, it's gonna have a massive impact on their future. Yeah. Yeah. And so that second concern of, like
24:09
oh, wait. This is the this is the real version of they took our jobs, like, Hey. It wasn't the immigrants.
24:15
It was the dorks building robots that took all your jobs.
24:19
Yeah, I do think that's true. In fact, it's kind of a paralyzing thought because it's just labor jobs, actually. It seems like the knowledge jobs are going first.
24:26
Like, for sure. Every two weeks, it's like, oh, video editors out of a job? Oh, Photoshop. The useless skill now. Okay. Well, lawyers
24:35
don't need them. Right? This is like it's like clearly the bread crumbs start to get laid where you're like, okay. Today, they haven't replaced all these. But what's gonna happen in v twelve of this, like, of this product. Like, this v twelve is just gonna do the job for doctors, for for, you know, for everybody, for for all these designers, for video editors, for all these things. And so It's kind of a paralyzing thought, and it reminds me,
25:00
they did this interview with Elon Musk, and
25:02
he was like, They're like, you're really worried about AI. Right? And he's like, yeah. I'm extremely concerned about AI. It stresses me out. I lose sleep over AI. And they were like, so why are you working on AI then? Why are you developing it?
25:16
He's like, well, basically, I think this is gonna happen either way. It's it is inevitable.
25:21
My mind told me that. He goes, and then I just had to decide, do I wanna see the AI apocalypse in my lifetime or do I wanna die and then not not get to witness it? He's like, I decided I'd like to witness it,
25:32
which is just a hilarious thing because I thought he was gonna say something else. I thought he was gonna say, like, you know, I can use this for good. And instead, he's like, I just rather see it in my lifetime than not see it. He's like, and then he's like, then I felt good. And I just operate. I just continued on. That's kinda how I feel when I see the AI, like, getting so much better. I'm like, dude, I don't know if any of my businesses will even make sense in ten years.
25:55
Well,
25:56
Alright. Carry on then. I know. What what am I gonna do about this? Like, am I really gonna go stop the progress of technology? No. Am I gonna anticipate this now in, like, change everything that I'm doing now
26:06
also doesn't really make sense. Let me just make hay while the well, while it's possible and see what I think that when it comes to AI and these robots, I think what it is is it's sort of like,
26:16
a squirrel preparing for winter, where between now and the next, like, ten years, the name of the game is just pile away as much as you can to get above this certain threshold to where you can make it through this winter that is never going to end. And so it's like between now and whenever this moment happens, if it's gonna be in five years or twenty years, the name of the game is just hurry up and, like, board up your home for the store to come just that way you're safe for it. And that you're just you're independent and you're alright. But once that once it, like, once ever that threshold or that time comes, a lot of people I think are actually gonna be screwed. I'm not gonna predict what's gonna happen because I have no no idea, but I just think that between now and then, the name of the game is just to make as much money set up your life in such a way where you can pile away all these nuts and you're just good and you're just good for the winter. That's, like, basically, how I've been living. I I don't know what the right mindset is. I think that's, for right right or wrong, that's also my mindset. It's just like, get it while it's time. Get it while it's time. Do do good while the game's good. Right? And it's like, well, I but I also am not, like, blinders on about this stuff. In fact, one of the name the name of my virtual holding company was inevitable outcomes. And I was like, well, why don't I just figure out what is inevitably gonna happen and then go invest in that? Because
27:30
that that that's just a better thing to work backwards If you realize that something is inevitable,
27:35
that's way different than speculating of, like, well, maybe this could be the case. It's like once you know something is inevitably gonna happen, then you just have to figure out which company is gonna win, which is actually, like, you've reduced the problem set down. Into which company is gonna win, and and when is the right time for for a business like this to exist. I I started this episode talking about timing because I think that these single purpose robots is gonna be the big winners in the next cycle here where, like, you have the knowledge work, like,
28:01
LLMs
28:02
chat GPT type of stuff, but then I think you're gonna see these specific robots. I think both of those are gonna be really, really successful. What you're not gonna see is the don't I think the human robot, I think it has to be further out than, than the asbestos, my my opinion. You know, there's also other versions of this. So we talked about the sweetgreen one. You know, the founder of Chipotle
28:21
has a new restaurant concept. That's all robotic first. What is it? Oh, and by the way, one other insight this week. Your guy said, he goes, the key is not to retrofit the restaurant.
28:31
He goes, what everybody tries to do is they have their existing restaurant, then they try to install a robotic arm that can try to use the existing restaurant layout because it's cheaper, right, to not rebuild a restaurant. And his quote was, everybody's trying to retrofit.
28:43
I'm different. I'm willing to blow the whole thing up.
28:47
Quote. I'm willing to blow the whole thing up, and he was basically saying, like, we just thought through from first principles.
28:52
If this was a restaurant that was not gonna have humans working on the line. It was gonna have robots.
28:57
How would you design every part of this restaurant? And then let's build those. And so they did two this year, the up and running. And now they're doing, like, ten more in the to scale this up. What's it called? And what do they serve? So is it called kernel?
29:09
So now, what's his name? Steve Els. Steve Els, the guy who created Chipotle. He's got a new one called Colonel. And Colonel is basically, like, some kind of, like, vegan bowl type of concept. And so you can see videos of it, where it's basically, like,
29:24
dispensing stuff into a bowl and then it's, like, shake there's like an arm that's just shaking the bowl just to mix it for you. It's pretty hilarious. The crossover
29:32
of vegans
29:33
and,
29:34
apocalyptic,
29:36
like, AI people. I don't that venn diagram, I don't think it's gonna cross over nicely. Like, you know what I mean? Like, Like, hey. Do you wanna go to this, restaurant that has no humans working ever? Cause fuck people.
29:48
Yeah. Do they do they sell hummus? You know what I mean? Like, that conversation is not gonna be as I
29:55
Plant based machine operated, baby. Yeah. Who's who's got that tattoo Nobody.
30:01
That's not a huge crossover, but it looks cool. So he has two or three of these locations.
30:05
Yeah. So I don't think this one's as figured out. It doesn't seem like it's as promising, but, there's a bunch of people doing this. Just SoftBank had poured, like, hundreds of millions into a pizza bot that I've talked about a lot, which is pizza is the big prize here. Pizza is a a much simpler dish. Like, yeah, it's a single single item. It's the most popular kind of like fast
30:24
casual type of type of food.
30:26
And
30:27
if you can do and it's also owned by chains. And so you can get if you can get Domino's
30:31
as a customer You're gonna get to go in whatever, like, you know, five thousand domino locations. And so whoever can build
30:38
a pizza bot that can make a perfect pizza. Every time, perfectly circular, perfect ratio of ingredients never calls in sick and work twenty four seven,
30:46
consistent.
30:47
It doesn't matter if that thing costs, like, you know, five hundred thousand dollars you'll be able to sell into all the dominoes, all of the pizza huts, all of the chain chain pizza stop shops in order to,
30:59
to do this. So I think the pizza bot is, like, you know, the big opportunity here.
31:03
I, I'm I like to think of myself as an independent thinking person. I like to live my life somewhat independently,
31:10
and I don't wanna rely on other people. I that's why I like to start my own companies.
31:14
When I went to Brett Adcox
31:16
warehouse or factory, whatever you call it, and he showed me, he was like, check this out. We're learning
31:22
how the Achilles on a human works. And we just got our humanoid
31:27
to move its foot, like, in a dorsal flexion way, and, like, We also, like, here's, like, this human need that we're studying. Now our guy over here is making the knee so we can, like, move. I saw this stuff, and I was, like,
31:41
Dad?
31:42
Can I come? It because it's, like, I was, like, can I should I drop everything? It just come and join you? Like, it felt like Adcock?
31:49
Yeah.
31:50
Sounds great.
31:51
It it felt like I was
31:54
in a cult. Like, I was meeting, like, Charlie Manson, and I was, like, Yes, sir. Who would you like me to kill? Like, I was just, like, so into this. I remember this going to my wife. I was like, you should quit your job. I was like, Sarah,
32:07
I don't I I'm not smart enough, but when you quit your job and go work here, I was like, I just was so into it when I saw this stuff. This is and and the and the cool thing about this topic is that it's happening this second right now.
32:20
And this isn't far away enough to say one day in the future. It's like, It's happening now and over the next couple months and couple years, we're gonna see it in play. And I think that's why this topic is very, very, really fascinating. The way, this is also the magic of Silicon Valley. Like, one of the reasons to be in the Bay Area is you will because he's a his factory is in the Bay Area. Right? Yeah. It's in, it's South San Francisco,
32:43
it's just north of Palo Alto. So you will go. You will either meet somebody You'll be at a dinner and hear something, or you'll drop by somebody's office.
32:53
And you'll see what they're working on and how they're approaching it, and then you will walk out.
32:57
Being like my life is meaningless.
32:59
Everything I do is absolutely meaningless.
33:02
I am a small
33:05
small pale dot on a giant rock, and I've met I mean nothing. And what I'm doing is nothing. Most people, that's a pretty, disempowering feeling. But for me, I wanted to be in situations like that. I wanted people who could basically, like, blow the ceiling off of what I thought was possible what I thought was cool, what I thought was interesting and be like, no. No. No. You thought that was interesting?
33:25
This is interesting.
33:27
And, that that happened the other day. I don't know if you saw these videos that were going viral. Some people write a hackathon down in, also in Sunnyvale or Mountain View somewhere doing, like, a there's, like, something called, like, the AI house or some shit like that. And during the hackathon,
33:40
during the hackathon, this guy walks in with this cool look of jacket and, like, disheveled hair. Like, he's been in, like, a craagenic
33:48
freezing tube for the last ten years.
33:51
And it's Sergei Brand. It's the founder of Google, just showing up at the Hackathon, and he gave an impromptu talk. Did you see this? No. What did he say?
33:59
Well, he goes up there and he's just gets really into details about Google's Gemini thing, like, about, like, bugs that they're fixing and stuff like that. Like, not like some grand visionary talk. He was just, like, yeah. Like, here's what the here's what I could do. Here's how it works. Here's how we're thinking about this. And he's just talking to a bunch of guys. Of them. This is hilarious. It went viral because there's a guy sitting there. The guy who asked the question in the clip, and his shirt
34:22
is just a shirt that looks like someone's body flesh Yeah. With boobs.
34:27
I
34:29
mean, I said with the Serious question. He's like, what is it about the, because, you know, like, Gemini got in trouble for being, like, too woke or whatever. Like, the thing wouldn't generate images of white people. And he's like, how come your you know, the LLM is not generating images or, you know, what? Which views on the policy of that? And nobody even calls that. Nobody, like, caregiver doesn't point. I didn't be like, dude.
34:48
Can we get a different question? Not the guy with the blue shirt?
34:51
I'm looking at him now.
34:53
It's just like a a guy who looks like me, just like a normal looking dude, but he's got a a flush cover, but it's it's tits. It's a it's he's got boobs on his shirt.
35:05
And
35:06
nobody references it. Nobody acknowledges it. It's just like, yeah. Carry on what? Live and let live. Maybe that's the San Francisco way. It's it's a very strange place, but it's strange in a way where kind of amazing things happen. Like, the founder of Google just showing up at your hackathon.
35:20
And, and answering some questions. This is pretty wild. I so I experienced this all the time in San Francisco when I lived there. There'd be so many times where it was, like, I've I've told so many stories about this where I've go to someone's office and there's just, like,
35:34
huge screens with porn playing, and it's, like, Well, they're just working on this this website and it's a and it's, like, they make porn. And so just like normal, we, like, normalize there. There's been other times where, like, I've seen, like, I've come to the office and there's people sleeping on their desk and stuff like that and just, like, weird shit that we would make fun of or
35:53
things on that TV show Silicon Valley where they, like, have nipp alert and, like, Like, I've experienced this in real life, and it's fun to make fun of.
36:01
But honestly, it's awesome. It is so awesome to, like, be around these freaks. You remember when hacker houses or, like, just getting going, I would go to these these houses,
36:10
and you would see that they didn't have shared kitchens. And, like, half the time, it was, like, just powders. Like, well, these guys just live off these pout they they just live off powders. So that's all they eat. And it's really started. You know, those
36:21
guys are just living in a house together. They're supposed to be working on other stuff. And then one of the guys was just eating only. He's like, I'm just gonna create, like, baby formula for adults, basically.
36:31
And he's just that became soylent. And that same guy
36:35
when, like, I forget what type of chip it is. When it came out, like, you know, like, these credit cards that we use now where you touch your credit card on a machine instead of inserting it and and and you could read the credit card. He inserted all of that into a small chip, and he surgically put it into his hand. And so he could just, like, touch and say, So, like, being around freaks like that, it's fun to make fun of them because it is silly.
36:55
It is awesome. I loved I loved being around that. I would see this stuff all the time, and I was just like, I'm so happy you freaks exist. This is the coolest thing I've seen. I felt like that all the time. Did you ever experience that when you'd go to these, like, hacker houses and it and they would be disgusting these guys. I just got it will it's just like triple bunk beds.
37:12
Yeah. And they're filthy.
37:14
But, you know, like, God, damn it. I can't even look down on this because one out of the one out of eighteen of you guys is about to become a billionaire in the next five years. I just don't know which one. Like, you know, whose neck beard is longest is, like, you know, who who's showered the least in this room? Like, I gotta, I guess, I guess, write you a check, I think. That I think that's the optimal strategy here. Like, I had a buddy
37:33
who worked at my company that would just do these things, these experiments, like,
37:37
his guy Quinn.
37:39
And Quinn had this, like, crazy, like,
37:42
Sideshow Bob style haircut.
37:44
And Quinn would be, like, he's in really in the games. And so he's playing this, like, little, like, word scramble game on iPhone.
37:49
And I think, like, his girlfriend had beat him in the, like, they were him and his girlfriend would play and this girl he was, like, flirting with or whatever would play.
37:56
And she beat him, And so he stayed up all night pro programming a robotic finger
38:02
and a camera to, like, solve this, like, word scramble thing so that he would never lose again.
38:08
I was like it's like these stories you hear about Kobe Bryant or Michael Jordan with, like, this, like, fierce competitiveness,
38:14
but applied in the absolute nerdiest way. And it results in kind of amazing things, and it's amazing energy to be around. I've been around a bunch of those people, and I love it. And it always makes me feel like, the first feeling I originally what I felt is
38:28
you're stupid, that's stupid. And then it's like, oh, I'm stupid. You're better than me. And this,
38:33
like, this is awesome.
38:36
Yeah, I I feel like that on a regular basis when I when I would go around these weirdos, and I love them. So we gotta also talk about this lawsuit. So Elon Musk sued Sam Altman.
38:47
And there's probably there's a good analysis out there about this lawsuit, but I wanna point out a couple of things. So here's my frame of this.
38:55
Elon is good at PR. He's not just good at engineering. He's not just good at business. He's not just a technologist. He's also kind of a master or PR.
39:04
I should have known this because Tesla runs basically spends nothing on TV commercials every other car company
39:09
spends, like, billions of dollars on,
39:12
you know,
39:13
like, all other car companies in total will spend, like, over a billion dollars on marketing.
39:17
Whereas Tesla runs no commercials,
39:20
yet Tesla's brand is huge, and it's because Elon is kind of a master of marketing, he's a master of PR.
39:26
And he does this with Twitter, he does this with with Tesla, and he's doing it with SpaceX,
39:31
and now seeing him do this with this lawsuit. It's the first time that I've ever seen
39:35
him do this in in public, but it's basically he's gonna lose the legal battle, but he's gonna win the PR war, I think. Have you ever been through a lawsuit? Do you know what the discovery phase is?
39:46
No. I haven't, but I'm familiar with the idea of discovery. So basically, I don't exactly know, what all the rules are, but I've, like, I remember there was a time I was gonna sue someone, because I thought it something didn't work out, and I and I I thought they had violated a contract, and I'd lost money because of it. And I was gonna sue them. Some guy didn't rack his weights.
40:04
Yeah.
40:06
Solving me. Broke code has been broken.
40:10
I can't let this stand.
40:12
And so I
40:14
was gonna do it. I hired a lawyer and they're like, alright. But they're like, you're probably gonna win this thing. You're not gonna win that much money. And I'm like, but it's the principle of the thing. Whatever. And they're like, okay. Cool. But do you know what discovery is? I was like, no. Tell me more. And they're like, we're gonna get access to your email, to your phone records, to your text messages, and they're gonna be able to see all of it. And in many cases and this is where I don't entirely understand how it works, but in many cases, that's completely public record. So if I file a lawsuit, that part is public record. And oftentimes, the the things they discover will be in public record. And I was thinking about it. It's like, Did I just text my wife the other day to let her know that I farted, but a little something came out? Because I think I did text that. And I don't want that in public. And that's, like, a three out of ten on the PC meter. Like, what's the nine out of ten? And would I be okay with them seeing that? And that's when it finally hit me. Of all the years of people telling me, don't write, like, offensive or crazy stuff on text messages or email.
41:12
That's why. It's because I just didn't want that embarrassing stuff to be seen. So I didn't end up going through with it just for that reason. And so in this lawsuit, we're gonna see all types of private conversations that these people have. I think Elon doesn't give a shit, but I think the average Joe would, including me.
41:28
Well, I yeah. I don't know what we should know more about this. I don't know how discovery works of, like, what parts get because not everything goes public, but Not everything. No. So it's like, what what's What's the where's the line? And how does that work? Because I think Elon has a lot to lose too with that. Like, nobody wants their shit out there in the public. And so But he does, but, like, when he when he when he texted that guy, the the former CEO of Twitter, he replied with all the stuff. I'm like, can we talk about this? An agent reply was, what have you done this week? Like, I I feel like in his mind, he thinks yes. He wants to do that. But in his mind, he's like, I'm gonna say stuff that I want the world to see. You know what I mean? It's, like, all part of the game, but in this lawsuit, there's actually, some stuff that's into that. I thought I found pretty interesting. So can I just, like, tell you what I thought was interesting in this? So he,
42:11
so it starts off, basically, the whole lawsuit starts off the the document And it says, well, humans have transitioned from a labor economy to a knowledge based economy.
42:20
And now knowledge is basically getting wrecked by AI. So first,
42:24
AI got so good. It could beat the best human chess players. Then it could find the best,
42:29
routes between two points, like best streets to take fastest route. Then it could be more complicated games like poker and Go.
42:36
And what it talks about is that Elon Musk saw this happening,
42:40
And he was, I think, on the board of or an advisor to a company called Deep Mind. And he's like, oh, Deep Mind's really interesting. He meets the founder of Deep Mind, this guy,
42:49
Dennis,
42:50
I don't know. I pronounced his last name. And he said he's basically like, what are your views on AI safety? Elon for a long time had been worried about AI safety. And he walks away, like, I don't think this guy is really, you know, concerned with AI safety. I think he's just trying to build really powerful AI. Okay. Whatever.
43:06
But he gets pretty worried because he finds out that Google is gonna try to buy DeepMind. And he's like, well, Google has the most data because they have emails, they have search, they have YouTube. So he goes Google has the most data. Deep Mind has a monopoly on AI talent,
43:21
and so they have the best AI talent And what the result will be is that AGI, when it eventually
43:27
gets done,
43:28
artificial general intelligence will be in the hands of
43:31
you know, the biggest monopoly than a private company out there. And he's like, okay. That's a little bit worrisome.
43:38
And he goes and he hangs out with, Larry Paige, the founder of Google, and he's telling him, like, Larry, I'm worried about AI. What are you guys doing about AI safety?
43:46
And he's like, basically, kind of the way Elon describes it, he kind of, to pass him on the head. He's like, oh, don't worry about that. We're not worried about that. He's like, well, you know, AI could take over, basically eradicate the human race. If it, you know, goes there's a path where this goes wrong, and he's like, well, that'll just be evolution.
44:02
And he's like, don't be such a specist,
44:04
Elon.
44:05
And he's like, specious. He's like, yeah, you're favoring the human species.
44:09
Oh my god. Yeah. I'm pro human. And so he gets really pissed about this. And he he says that this is when he stopped talking to the CEO of Google. So he after that, that's, like, the trigger according to Elon. And again, I think a lot of this is Elon crafting a narrative. I'm not sure. I'm not saying that these things didn't happen, but I think he's very good at creating an origin story and creating a bad guy and then pay to me himself. Oh, what do you know as the hero? And so he here he goes. So now what what he does is He teams up with Sam Altman. So Sam Altman, who's the CEO of OpenAI in two thousand fifteen wrote the following. He goes,
44:42
the development
44:43
Of super machine intelligence at the time that's what you're talking about. SMI, super machine intelligence.
44:48
It's probably the greatest threat to the continued continued existence of humanity.
44:52
There are other threats, but they are unlikely to destroy every human in the universe the way SMI could. What a great what a great opening
45:00
sentence.
45:03
Alright. Could you, Sam?
45:06
So
45:07
Sam is worried about this. Elon's worried about this. They get together.
45:11
Because oh, I then there's the trigger. Google's buying deep mind. I didn't I didn't know this.
45:15
The in the lawsuit, it shows that, Elon then went to the PayPal cofounder, this guy, Luke Nosek,
45:21
and try to quickly, like, emergency raise five hundred million dollars to buy DeepMind instead of Google, just to keep it out of the hands of Google. And he fails.
45:29
He can't pull the deal together in the in the in the last minutes. Deep mind gets bought by Google, and now now Elon's really worried.
45:37
And so Sam Altman then goes to Elon and says, hey, why don't we create a non profit AI lab to catch up with Google? And, actually, the origin of this was to do it inside Y Combinator where Sam was currently the president. So how did you know that that the the origin of opening I was actually supposed to be inside of YC And he called it, what if we created a Manhattan project,
45:57
for AI? And he's like, what if YC did a call to arms and was like, hey,
46:01
We need to develop this technology to, basically, we need the anti Google. Like, we need to get the best people we can to start working on this as a research project. What he said originally was we'll give them equity and YC as their upside.
46:13
And so they can work on this as a nonprofit. That was the original idea. I think he said that, like, no person will have control over this thing that we're making and and said they were gonna get equity and YC. That's how they're they'll He could be nonprofit. The how do we pay these top AI researchers? Because Google's gonna pay them a shit ton. Well, we'll give them equity and YC. YC's are are huge, you know, cash cow. And so that'll solve that problem. And he's like, we'll have a board of directors, me, you, Elon, and then we'll pick three others. And, like, if it's he he says, this is the, like, initial email. He goes If it's not obvious what the right thing to do is, then the five of us will decide. Right? That's the governance idea. Okay. So
46:50
That was the original idea. It evolves a little bit. By the way, Elon's reply to that email was great. And so Sam wrote, like, a fairly casual I mean, he used the, verbiage, like you said, those sentences that are, like, a big deal, but it was, like, a casual, almost email with, like, five points
47:07
And Elon's reply was agree on all. Like, that was it.
47:11
Right. I have moved straight to the end. So, yes, agree on all. So they start, and they're like, cool. We're making this nonprofit.
47:19
And it'll be open source. Elon comes up with the name OpenAI.
47:22
He's like, let's put it in the name. This is open source AI. And in the next five years, Elon contributes forty four million dollars to this nonprofit.
47:31
And
47:32
they developed some pretty interesting techs. So chat GPT now GPT four,
47:37
it's it could do it like an it's in the ninetieth percentile at the lawyers like the bar exam. It's in the ninety ninth percentile,
47:44
for the GR, the verbal section. It's a even a seventy seven percent at the advanced Smolier test, which I don't even know how that would work. It's like the Smolier, like, at the wine test.
47:55
And so he's trying to, in the lawsuit, argue that it's already AGI. Which it's not, but but he's trying to make that argument,
48:02
because
48:02
his whole his whole legal case rests on this idea that There's been a breach of contract, and they breached the founding agreement. But there is no actual founding agreement. What he's calling the founding agreement is like what we call the gentleman's agreement. It's basically like a handful of conversations,
48:17
emails,
48:19
and then just like the initial founding, like, in articles and corporation of the of the
48:23
company. He calls all of that together the founding agreement. He says they breached that, but it's not actually a,
48:28
a, a agreement.
48:30
I'm a moron.
48:31
And I don't know anything about legal stuff, but
48:34
I think that that actually counts as I mean, there's been I think there's precedent where that matters So but in this case, basically, he has to establish that there is a founding agreement and that it's been breached. But if you go look at the founding agreement, like, those emails, like, One time they're talking about we could give equity and YC, that's just like brainstorming. That's not like the agreement. Then later, they they do have a little bit more formal. It's like, okay, it's gonna be a nonprofit. It's gonna be whatever. But in the articles and corporation, it specifically says, like,
49:00
we will open source this,
49:02
you know, whenever,
49:04
you know, whenever we deem appropriate.
49:07
Yeah. It's vague. It's vague. Right? It's like we will really we will develop this for the benefit of humanity.
49:12
We're definitely doing this maybe.
49:15
Someday.
49:16
Exactly. When we want. And so they basically left themselves, like, complete latitude to do whatever the hell they and that's what they did. Now what I think is interesting. So Elon says a couple other things. So
49:27
one, he says GPD four is already so good at, like, those test scores. He goes, And now they're developing this thing called q star. Have you heard about q star? No. What is it? So q star is a reportedly, like, the new
49:39
What what what this is not reportedly rumored to be the thing that they open as working on that is, like, next next level after GPT four. So it's, like, Instead of GPT five, they're using some method called q star, which they think is like a leap
49:53
in the intelligence, which
49:55
is really exciting because even GPT three to GPT four, GPT two to GPT three, like, those are, like, pretty substantial improvements,
50:02
but not, like, a huge leap whereas hustar reportedly is, like, what the huge sleep is. There's rumors where Sam Altman,
50:09
came out and said, you know, like,
50:11
I was in a meeting, like, a debt we did a demo where we had an internal meeting the other day, and I saw, you know, I saw something that, like, blew my mind. Like, basically, I saw something that just, like,
50:20
was awe inspiring.
50:22
And I was like, what? Well, what is it? And he hasn't shared any details. This is also right when the that big coup happened. They fired Sam Altman. I thought of the one of the rumored reasons was when this could be total bullshit. Wait. We don't know. This is one of the rumored reasons was
50:36
The the board saw that how powerful q star is and they're worried Sam is way too fast and way too loose loosey Gucci way too aggressive And now this weapon is getting too too powerful. It's kind of the the idea there. So anyways, he talks about the Elon talks about that in the lawsuit that they're developing q star and whatever. It's kinda like confirmation that this thing exists.
50:55
Although
50:56
the fun the legal language he uses in all this, he goes,
50:58
at the every start of every sentence, every paragraph, it says,
51:02
using knowledge and belief.
51:05
And, basically, it's like saying, like, we don't have evidence, but, like, I know some things, and I believe this
51:10
to true. And then it says all of the claims. So who knows? It's kinda like Elon's claims. A couple other big pieces of this. So I wanna read you some, some Elon emails that that did get surfaced in this. So
51:22
one of the key things that happened early on was they had this dinner. And at the dinner at Sam Altman, it's Greg Brock,
51:28
It's Elon, and then it's this guy, Iliya. Iliya was the lead you know, one of the leading deep deep learning researchers at Google. And he became the chief scientist of of an AI. Those are the big
51:38
Turning point was he Elon recruited Elya away from Google to Open AI. This is when Larry paid stopped talking to him because he was so pissed that you recruited my, like, top guy, and you declared war, basically.
51:50
And recruiting him was, like, a herculean basically getting him to leave Google and Google was offering him, like, a blank check. You're in a fucking movie. And, exactly. Well, this will be a movie for sure.
51:59
So then Google starts offering all of their talent, like crazy lavish offers. They just up the ante like crazy. They're like, we're gonna price anyone out of this market. How much? Do you know? A million, like, I don't know, like millions of dollars per per employee minimum for some people ten, fifteen million dollars,
52:15
a pay package annual.
52:17
And so it was like, okay. Well,
52:19
Sam and Greg email Elon saying, hey, it's getting really hard to recruit.
52:22
You know, we have to think of a strategy here. Google is doing this. Elon replies. He goes, We need to do what it takes to get top talent. Let's go higher. If at some point we have to revisit what our current people are getting paid, that's fine. Either we get the best people in the world or we will get whipped by deep mind. Whatever it takes to bring on the Ace talent is fine by me. Deep mind is causing me extreme mental stress. If they win, it will be really bad news with their one mind to rule the world philosophy.
52:47
They're obviously making major progress, and they should given the talent, level of talent over there. And so he basically gives them the green light. Like, I'll keep funding more. Go get the best talent. And so in twenty sixteen, he gives them fifteen million. In twenty seventeen, he gives them twenty million. That five year period gives them forty four million dollars in total. Now the problem is,
53:06
Elon, the biggest backer in the original backer,
53:09
says he doesn't own anything in Open AI, the for profit company.
53:12
And the this is the strongest part of his case. Here's what he says. He goes,
53:16
How is it that you can create a nonprofit company that is supposed to be nonprofit and open source?
53:23
Raise millions of dollars from people. You take fifty million dollars from me. You hire the best people in the world, and you you write down that this is your charter.
53:30
And then you end up a closed source
53:33
maximum for profit company
53:36
owned by the most valuable company in the world, Microsoft owned forty nine percent by them.
53:41
How is that possible? And he's like, two things. He goes,
53:45
if he goes, if it is possible
53:47
to create a nonprofit, hire people, and basically fund all of the research on, you know, basically, like, in a tax donation way where you for every dollar you're putting in, you get fifty cent rebate back from the government.
53:59
He's like, if that's possible, and then you could just flip to a for profit when you make breakthrough technology,
54:03
why is everybody not doing that? Why is that not just the default for every single technology company in the world? Either it's illegal or everybody should be doing it. It can only be one of those two things. And I love this argument from Elon. I think this is a tremendous argument And there's no answer to that question yet. Either this is illegal
54:20
or every single company in the world is gonna do this. Why would you know?
54:24
We you and I are not. I wouldn't I'd say we are fans of Elon. Elon's work, but we're we're lukewarm on him as a human being.
54:32
And we are only hearing one side of this story.
54:36
But in the last ten minutes,
54:38
only knowing what I know just from those ten minutes, he's right.
54:43
So then then he gives this analogy. Again, master of PR, he goes, it's like if I donated to a nonprofit charity that said, we're gonna save the Amazon rainforest.
54:53
And then that charity took all our money, and then they created a for profit logging company that was just cutting down trees in the rain forest and selling the lumber.
55:02
That's what's happened. That is Open AI. And so He says it is a de facto subsidiary of Microsoft. And he goes, it includes this quote. And this quote, I think is gonna become very problematic. I predict. It's gonna become very problematic for Microsoft in the future.
55:17
So it's the Microsoft CEO
55:20
saying, what remember when all the open AI drama was going down and Microsoft stock price started to tank because they were like, oh, shit.
55:27
You put you've given ten billion dollars or whatever to open AI, and they're about to, like, collapse
55:33
This is bad for Microsoft, and they're gonna go create, like, a new company that you don't own. That's bad.
55:38
And so then he came out and he said this publicly. He goes, If OpenAI disappeared tomorrow,
55:43
we have all the IP rights, all the capabilities, we have all the people We have the compute, we have the data, we have everything. We are below them, above them, around them.
55:55
The the CEO of Microsoft said that. Wow. Yes. Exactly. And so that also weakens the case that OpenAI is this, like, nonprofit non controlled by Microsoft entity when he came out and said that we are below them, above them, and around them.
56:10
God damn. What a what a what a phrase.
56:14
Dude, this is,
56:15
riveting. This is riveting.
56:18
That's me when I if I go to Chick fillet and I order nuggets,
56:21
I'm below them. I'm above them. I'm around them. I have them. They are mine. Like, dude, that's it. What a what a phrase.
56:28
The, by the way, the the way they call this the Manhattan project
56:32
No. The best. That is the best. That is so good. All of this is just epic. Can you imagine these nerdy,
56:40
AI engineers who were probably, like, bullied and were dorks throughout all the years. Now they're, like, Oh, me? You want me? No. Boys. Come on. Let's be civil about this. They're just getting fought over, and we're talking about fifteen or twenty million dollar a year packages.
56:56
That's insane.
56:57
Everything about the story is insane. You're right. This needs to be a movie. If you're making the movie or you wanna make the movie of this, get in touch with me. Let's let's create the movie of this. This is gonna be like the social network movie all over again.
57:08
An epic photo collage something like that would also suffice too. Like, just, like, some some neat photos with captions that tells a story. I'm into that.
57:17
I wanna leave it with this. So Darmesh shot. Darmesh is the, founder of Hubspot, and you and I have talked to him a bit. He's on this podcast all the time. Hold he he's doing this thing on AI that I want him to come talk about. Hopefully, he will. He told me something that I tell myself all the time when thinking about this, which was basically Dharmesh is super into AI. He goes a lot conferences on it, like insider conferences. See, I would think that of he has a a good grasp of it. He has said it's not probably gonna be as good or as bad as you think it's going to be. And him saying that in a calm, like, fatherly voice to me, that's, like, been the one thing that I've I've been thinking about of, like, Sam Altman. I think he's a great guy, and I love Sam.
57:56
I don't know him, but it just has, like, a fan of his.
57:59
I think that the all these guys in this game are also brilliant marketers
58:04
and brilliant business people and brilliant at the skill of programming and making this stuff. But I also think that they're brilliant at PR marketing.
58:12
And I have to remind myself constantly
58:15
that it could be good. It could be bad. It's probably gonna be right in the middle, though, of,
58:20
of what the reality is gonna be in the middle. And these guys are just excellent at telling their story. Do you agree? Yeah. And we should say, you know, Elon is create a direct competitor to
58:30
Open AI. Right? Like Grok,
58:32
he has he has XAI,
58:34
which is, like, his,
58:36
his AI company. It's his AI competitor to Open AI. And he has GRock, which is the competitor to chat GPT. That's like the product name. So
58:45
he obviously has, you know, interests in this. And, I think, you know, one thing that's clear out of this is it's definitely mission driven. Like, in the same way that, I used to think when it was like, oh, Jeff Bezos is doing a rocket company. It's like, oh, is this like a giant dick measuring contest between, like, billionaires. Like, just the last level of the game is you say, I I win Earth. Okay. Let me go win Mars now. Right? Like, kinda what it seemed like was, oh, I'm gonna build a bigger rocket than you.
59:10
But then you read that, like, even twenty five years ago,
59:14
Bezos was, like, really obsessed with the idea of space travel. And, like, has been a giant fanatic about it and and been been trying to donate to it and trying to advance the cause in that. And that now, like, even with these companies, he's in all the engineering meetings, the plan meetings
59:27
every week. It's like, oh, okay. This wasn't what I thought it was in the same way. By the way, it could still it could be that, and that's okay. It could be mission driven and ego driven and profit driven, and that is okay to have all those things set up. That's a better way of putting it. Yeah.
59:42
Like most things,
59:43
it's accelerating because of multiple causes, multiple there's multiple things, multiple tailwinds pushing it. But I would say, like, the hather in me wanted to discredit it, but
59:52
the realist in me when I go back and read old interviews or watch all videos before Amazon was super successful.
59:58
It, it becomes clear that actually it was at deep interest. So similarly here,
01:00:02
even though I think it's part of it is he's trying to
01:00:05
bog, open AI down with a lawsuit and spin the PR battle in his favor,
01:00:10
and he has a direct competitor. It's clear that opening, I wouldn't have even would not have existed. It would not have been successful. Had he not genuinely had this fear and concern and belief about what needs to happen with AI long time ago. So, you know, that I think that's that comes out very true in this. There's a paper trail of evidence.
01:00:27
What a fucking pod?
01:00:29
I feel
01:00:31
amped and scared,
01:00:33
excited,
01:00:34
nervous,
01:00:35
I feel all types of emotions. Great job. It's like a great first date. It's like, right? You are just supposed to, like, trigger like adrenaline or, like, the fight or flight response in people, and that's what bonds you. I feel like that just happened to us. Yeah. Except my version of that was, like, on our first date, we went on a motorcycle. But, now we're talking about, like, the world may be ending in twenty years.
01:00:54
Good job. This is usually the good job. I've got a court
01:00:58
case to tell you about.
01:01:02
Dinner? No. We're not doing dinner.
01:01:04
I okay. So you made it this far to to folks listening. Go to the YouTube and tell us in the comments.
01:01:11
Do you actually think this is gonna be horrible? Do you think this would be great? Or is it what Dharmesh said where it's somewhere in the middle where it's not gonna be as good or bad as we think? But Sean,
01:01:19
Bravo. You did a good job. This was a great good pod, and we'll end it there. That's the pod.
00:00 01:01:45