This Week in Startups - All-In E14: Salesforce acquires Slack, DeepMind’s AlphaFold breakthrough, Trust Fund Socialists & more

Episode Date: December 4, 2020

Follow the crew: https://twitter.com/chamath https://linktr.ee/calacanis https://twitter.com/DavidSacks https://twitter.com/friedberg Follow the pod: https://twitter.com/theallinpod https://linktr.ee/allinpodcast

Okay, besties are back. Besties are back going around the horn. Rain Man, David Sacks, calling in from an undisclosed location, suffering through two Code 13s in one lifetime. And David Friedberg is here the queen of Kinwa, spacking everything in sight, living the life, calling in from a nondescript Ritz-Colton room, it appears to be. And of course, the dictator himself, Chmoth, Polymouth. hoppetia cackling like a fool. Welcome back, everybody. This is what you pay for with your subscription to the all-in podcast brought to you by Slack. If you didn't own Slack shares, raise your hand. It's been an incredible week on a number of levels. We're going to talk this week about Salesforce buying Slack, Trump and Section 230, the Coinbase, the Coinbase, the going Coinbase saga. Freeberg found some interesting science that could save

humanity. And of course, the trust fund socialists in the New York Times who hate their parents for giving them money. Let's start off. Let's start with the most important thing. What is that shirt under shirt combo you're wearing? I mean, look

you have buttons on buttons. It's incredible. Did I break the layering rule? You can't. School us. If you're going to layer properly, you can have only one layer of buttons, but to have two layers of buttons, that's not how it works? No, Jaycal went in and got an almond.

Layers are for players, not me. No, he got like an almond milk cappuccino, and he's like, I like how that barista dress is, and I'm going to wear that from now on. Wait a second. Can I ask a technical question? Can I have buttons? I can't have buttons on buttons, but can I have buttons and then a zipper up like with the... No, you can't do that either.

Listen. Chamatha. Chimatha's had a weird aversion to buttons ever since he spent the time. time in Italy. Was he button-shamed in Italy? I was a little button-shamed, but I'm looking at Sacks has buttons on his collars, which just makes no sense. Sacks is wearing the same Brooks Brothers shirt that he graduated high school in.

At Brooks, he owns 17% of Brooks brothers at this point from the number of blazers he's bought there. All right, let's get off to it. We've insulted each other. I don't think Freeberg's taken the brunt of anything yet. Anybody have any chop busting they want to do with Freeberg? just sort of built in. No,

Freeberg took the tablecloth that I used for a picnic in the summertime and made it into a shirt. There is. You know, you have to be frugal at this time. And also,

Freberg cares about the environment. He's not going to just let a picnic. It was a hemp-based tablecloth. And so I knew it was going to get taken and stole. I love how I choose to spend my time with you guys. It just pays off. Here we go.

All right. Can we kick this off? All right, let's get it off with our advertisement for nobody because tomorrow will not let me make any money off of this podcast. And thanks again for the suggestion that we launch a syndicate with no carry. Now a bunch of dipshits on Twitter are like, hey, when is the all-in syndicate starting? I'm like, never. I need to make a living. I need to get my beak wet. I'm sorry, but I think an all-in syndicate would be super, super disruptive and cool. I'm totally fine with running it as long as we can have

the 20% carry and I'll manage the whole thing. We got four people on the call. We each get 5% carry, but we got to make a living here. Not everybody's made it, Chimov. Not everybody's got Spax A through Z or had all of their slack shares bought. I think we'll kick it off with that. Chabath, we saw this week. In fact, just two days ago, Salesforce in a record transaction for a SaaS company, I think it's the highest ever paid for a SaaS company, $27 billion, $27.7 billion for Slack, which has only been public for just over a year, I think. You were, I think, did the series B in Slack right after they did the pivot at Social Capital. I don't know if that was in front one or two, but Phil Helmuth keeps talking about it. It was the series B in tiny spec, but it was the series A in Slack. And there's a really important story, which is that myself and Ray Co, who's my partner in

Social Capital, we've worked together now for my gosh, I think it's probably 15 years, wrote a really great memo justifying the investment in Slack. And it had to do with one thing and one thing only. We ignored the revenue and ARR. I mean, it was fine and nice and good. But the single biggest thing that we were attracted to was something that we looked at, and which was called intercompany edges. And even back in 2015 or 16 when we did this original investment,

there was this dynamic where people across companies were communicating via Slack channels. And I was completely stunned by this idea because that was effectively a substitution for email. Because the only way you communicate across companies today is by email. You know, David is at craft ventures.com and he emails me at social capital.com and I email jason at inside.com. That's how we communicate across businesses. Except now all of a sudden, you could be messaging and having a much more real-time interface. That to me was incredibly disruptive and it justified the entirety of the real forward-looking

investment thesis. Now, fast forward five years later, and these guys have more usage on the way. on a daily basis than Facebook, which is stunning, because these guys have 10 million DAUs and Facebook has 2 billion. So it just goes to show you the quantity of traffic and the volume of information and theoretically, productivity that's happening on Slack. And so I'm not sure what Salesforce bought.

I actually think that you can make a case why it's a shame that it got bought, a very strong one in fact. But what they did get, whether they know it or not, is an intercompany edge effect, which is the most disruptive thing to email. And in the hands of Salesforce and that sales team, I think it has the ability to really be a very disruptive force for good in enterprise software. All right.

So, Sacks, this is a natural passing of the ball to you on the baton because you did Yammer, sold it to Microsoft for a billion dollars. And obviously Slack was the mobile successor to the desktop version of Yammer. you got a lot of your fingerprints all over this. But the fact is, you did a tweet storm about it. Slack is an unbelievable success. Stewart is a great founder.

You know, he sold the first company Flickr for $30 million. This one for almost $30 billion. So that's pretty nice. But there was one failure. And you pointed out in your tweet storm. Explain what the one failure, if you could pick out of the hundreds of things,

thousands of things they did right. There was one thing they did wrong that, to Chmott's point, would have resulted in them. remaining an independent company that could have become worth more than $27 billion. Yeah, it was it was a slowness to embrace the idea of enterprise sales. And by the way, let's put this in context. I mean, Stuart and the Slack team did a phenomenal job, $30 billion exit,

seven years of just about flawless execution. So I don't want to, and also, you know, I was an investor in the company. So thank you to Stuart for letting me invest. I'm definitely don't want to. sound like an ingrate or a critic. I mean, they just, they did a phenomenal job. But if you were to nitpick just one, one little thing that I think they could have done faster,

it would have been embracing enterprise sales. The big learning from Yammer, we learned this at Yammer from 2008, 2012 is that enterprises don't self-serve, right? They don't self-close. Bottom up, SaaS products are phenomenal for generating top of funnel, basically, generating leads, but you have to have salespeople close the deals. And enterprises don't just kind of pull out a credit card and self-serve. They need a salesperson. And I think there was

something in the DNA of Slack that actually I see really very commonly in the DNA of sort of producty SaaS companies, producty SaaS founders, which is they kind of have a reflexive dislike or distaste for sales. and they resist the idea of sales and they want to believe that they can just be entirely product driven. And what I see across the board is they all come to the same realization that we had a Yammer, which is we have to have a sales team. And I do remember, you know, back in 2014, the whole Yammer sales team was basically rolling

off because of, you know, Microsoft acquired the company in 2012 and that was an integration period. And by 2013, 2014, they were all looking for jobs. And I remember, you know, my former CRO, I think was interviewing at Slack. And it would have been such a perfect thing for them because he had just learned all the lessons of how you layer on kind of an enterprise sale on top of a bottom up product. And they just weren't ready to make that higher yet. And so look, if you're going to nitpick, look, $30 billion outcome, no one's criticizing. But if you were to nitpick, you know, it's an A plus regardless.

you know, this would be the one thing you could, you could say. Well, congratulations all around to everybody involved, especially Phil Helmuth, who was an LP in one of Chimaut's funds. So if you need insights on Slack or anything inside information, you can just follow Phil Helmuth on Twitter at being the greatest or I am the greatest or I'll always be the greatest. One of those Twitter handles is his. But Jason, I mean, you basically came to the same conclusion in your emergency pod, right? I mean. I did.

I hadn't seen your, I think your tweet came after the emergency pod. But yeah, it just seemed to me this company, unlike Zoom, should have been able to grow quicker. And if you look at their numbers, they had 87 companies that were spending over a million dollars.

You put a rabid sales team on that product and they go in like Benioff does with his sales team. I mean, he was just hyper aggressive at just putting huge numbers out there and saying you have to pay us this much money. So much so that I don't know if you remember Elon getting into a public spat with him where he's like, Salesforce is horrible software,

get it out of the organization. He basically banned it because they came to him with the bottom up people using of Salesforce and said, hey, you owe us this amount of money. And Elon was like, F you, banned forever from inside of our organization. We'll build our own software. We don't need it.

And they didn't have somebody and Stewart didn't have that DNA, I think, to say aggressively, we need to charge what this product is worth. you saw that in, I think, one of their strengths and weaknesses, which was they only build you for people who were actively using the product. Now, that's a beautiful, awesome feature. It makes you not scared to use it. But on the enterprise level, I mean, that seemed to be like maybe one of those non-cutthroat things that maybe we're holding them back. David, you have any insights on this?

Or should we go on to Alpha Fold? It's really important to remember the mechanics and the game theory around M&A, especially, you know, big game hunting when you're doing 30 billion dollar acquisitions. It's also kind of true at billion dollar levels, but less so. But the bigger the acquisition gets, you have to remember that there's an asymmetry of information between buyers and sellers. And the question is, who does the asymmetry favor? Right? Because you could look at this acquisition and say, wow, Salesforce is crazy for spending $30 billion.

And somebody else may say, wow, Slack was really stupid for selling it for $30 billion. The reality is that I think that there was asymmetries on both sides. I think that what Slack probably saw, and I don't know because I've been off the board now for more than the year, but I think what they saw was, as David said, just a level of sophistication and scale and ability to cross-sell and up-sell that was needed for enterprise scale. Either you overcome it with precision and speed or you overcome it by going the same pace as somebody like Mike. Microsoft, but with an equivalent product portfolio. So that's sort of one realization that Slack had. But in the case of Salesforce, what they probably had was a realization that they couldn't go wall to wall inside of a customer because they didn't really have a product that was

useful or usable to every single individual inside of an enterprise. And so both of those two things create asymmetries. There's a level of fear inside of Slack. and there's a level of fear inside of sales force. Both of them are about the fear of disruption. And then the question is, who gets the better of the other person in the middle of the acquisition, right?

So the deal could have probably gotten done at, you know, $22 billion. It probably could have also gotten done at $45 billion. And that's, again, to a combination of how well you play poker in that moment, right? Who blinks first and the quality of the bankers? This is like two people having top pair on a very textured board. It's like, and they're just raising versus each other.

Yeah, it's who up places them. Because like, it's similar also to like how Microsoft bought LinkedIn, right? Because if you think about what happened in LinkedIn, if you remember when that happened, it was almost to a T very much like Slack. LinkedIn had a one bad quarter. They got decapitated by, you know, and I owned it at the time in the, in our public fund. It got decapitated by like 50, 60, 70 percent. I mean, something insane for missing numbers by like a few pennies, okay?

And all of a sudden. it took a lot of the wind out of their sales internally. It didn't change the user momentum at all, because the users that were signing up for LinkedIn didn't care what the stock price was yesterday, today, and tomorrow. But it all of a sudden created a fear. And I think Microsoft was able to exploit that fear. And within a year, this company was bought for $25 billion. Not dissimilar, you know, Slack had a hiccup and they got re-rated. The stock bounced back. But I think that if Salesforce was smart, they probably created, you know, sort of like a white knight kind of bid that said, listen, you need enterprise scale and the ability to cross-sell and upsell,

I can give it to you. And Slack probably said, listen, you need to go wall to wall, so I understand why you need me. And, you know, the price is what the price is. If you look at the pricing, right, so Slack, normally the way these big M&A, you know, public company M&A deals get done is the board has to approve the price. And they have to say, this was the right deal for us relative to other options. And one of the ways you assess that is you look at where the share price has been historically. And if you're getting a premium to where the share price has been historically, let's say 30, 40 percent higher than it's ever been, then the board says, great, that's a good deal. We should take it because we've got a long

way to grow into that value. In this case, the deal was done not at a very high premium to where Slack traded just in the summer. Is that right? Chimotsu, it looked like it peaked. It's basically, if you look at the fully deluded. It's a 10% premium, right? 10% premium, yeah. 10% premium, yeah. We opened the direct listing at 40 or 41, and then this was at 45. Right. And so there clearly was a sense of weakness from the board, which is, I think, why the

Salesforce stock traded down afterwards, because if they were willing to sell at that small of a premium, the forecast internally is probably feeling not that strong. And then people translate that into, hey, Salesforce bought something that's not that strong. You know, there's something a little bit amiss. But obviously, to your point, they're missing a lot of the cross-selling and the synergy that will arise. I think it's a huge slam dunk acquisition. And I go back to this idea of intercompany network effects. I think they exist and I think they're real.

And I think that the Slack product team's ability to innovate around that was not as fast as it could have been, but it was still very unique. And I think it was a true moat. and the tragedy is we won't see what the terminal value is if they were left alone to execute. And in this weird way, like I've always struggled with why Microsoft's was so overly obsessed with Slack

because if you looked at the team's product, it was much more directly competitive with Zoom. And to this day, still remains much more directly competitive with Zoom than Slack. But, you know, there we have it. And if you look at the revenue, Slack was doing 800 million run rates. So anyway, rounded up to a billion.

And you had Salesforce at 20 billion. So 5% revenue to revenue. And then they got 10% of the company. So in that way, if you look at it on a percentage basis, which is, you know, how you might look at the Facebook, Instagram, and WhatsApp is what percentage of the existing entity did they get? Right. The size is growing about 60% a year and Salesforce is growing about 22% something like that.

The other thing is the president of Salesforce. is Brett Taylor, who was our CTO at Facebook, who I worked with. And so I think Brett also understands network effects really well. And, you know, by the way, in this interesting twist of fate, Benioff was the underbitter, I think, for LinkedIn. And so, you know, we've seen Mark around the hoop on these, you know, social network, network effect, business tool acquisitions before. And finally he got us.

Also Twitter. He was running around the basket with Twitter. And then they also brought his name up for TikTok. which made no sense. So I think Benioff is just looking at this like if Google and Microsoft and Apple are too scared to buy things because of antitrust, well, I'm under the radar of the antitrust trillion dollars. So I'm the only game of town, right? He's under the radar because he doesn't have a play in this sort of communication

or collaboration space. And so therefore there are no antitrust issues. If Microsoft were to do it, it would definitely be scrutinized because you could argued that they're adding to their existing dominant market share and in collaboration. But Benial's dream has always been, at least since he launched Chatter to compete with us when we were doing Yamers back in 2010, 2011, his dream has always been to have a product that could get him onto every seat in the enterprise. His current product set is departmental.

I mean, you've got kind of the CRM product for sales and they've got the support. cloud for customer support and they've got the marketing cloud for marketing. And so he's gone department by department, but he's never really had a sort of pan, like sort of cross company. Yeah, something that the entire company would use. Like a central login system, right? And Slack is that central login system. But when you, when he came up against you, it was very, you know, Benny Off, you're friendly with Benny Off. Benny off came at you so hard. He threw three or four hundred engineers at chatter. He took out full page Wall Street Journal ads.

He tried to poach your people. He tried to make the product free. He made it personal against you after you would not sell to him true or false David Sachs. I don't think he made it personal, but it was definitely a perfect. Did it feel personal? Did he hear your guys? No, no, no, no.

He did. I understand what he was trying to do. That's your way of saying. No, I mean, if we had sold to Salesforce, like we, we ended up. So what I would say is, yeah, we got in like a very, it was a very competitive situation. He didn't beat us. You know, I, I, what's that? He felled.

Does that product even exist? Yeah. It's sort of like a feed inside of the, the CRM product. It didn't really succeed as a standalone collaboration product. And so we won that battle, but it definitely, I would say it scared us enough to sell to Microsoft. Because, you know, the, what did he offer you? We were about to enter a new stage of competition.

So here's what happened is he launched his product to kind of be a clone of Yammer inside of Salesforce. But he was initially charging $15 per seat. We were charging like five. And so they massively overpriced it. And then they were on this like slippery sloper. They kept lowering the price to compete better with us. And then finally they realized that they should just give the thing away for free as a strategic move. And that was when we decided to sell to Microsoft is we didn't know.

We knew we had a better product than chat. but we didn't know how it would go if we were up against a free chat. Tell us honestly, how much did he offer? What was the meeting like where you made you the offer? We, yeah. Take us to the more. Yeah. So here's, I'll tell you the backstory.

I mean, this hasn't been publicly revealed, but. Here we go. In service of the all-in podcast. Go ahead. Go ahead, David. Get us some readings. Yeah, to try and services trying to get us from number three to number one on the charts. No, you know, it's funny.

We launched Yammer at the TechCrunch 40 conference that Jason, as you know, you were the co-founder of. And Benioff was like a judge. He was a panelist. And he was raving about it. And you could just, you know, from the moment we launched, he was raving about it. You could see the light bulb go off with him. And he realized that like social was going to be. It was, you know, at the time, obviously social was big with consumer social networks, but

he saw the potential of social or collaboration inside the enterprise. And so, yeah, I mean, like, I think a year later or something, they were interested in buying the company for around $250 million. The big issue for them, though, was that Benioff had a bunch of like engineers who wanted to build it in house. And so they, they actually, I don't know what would have happened if, if they, you know, didn't want to build it themselves. but basically they vetoed doing a deal.

And so they ended up building chatter. And they threw the 300 engineers at it. And they basically spun their wheels for a few years. And anyway, it turned out to be much better for us because we ended up selling the company for five times as much to Microsoft. You know, if we had sold to Salesforce in like 2010, it would have been a much smaller deal. But yeah, I mean, he was very interested in it from the get-go.

All right, folks. So you have a breaking news in the background on what actually happened. Congratulations to Stewart and the team. I want to ask a question, Chimov and Sacks. Did you guys keep all of the shares you originally invested in to the exit here? Just to set the context for folks, you know, you invest in a company. It's a small startup. It's equity for $30 billion.

Yeah. For every share that I owned, half of it were half. No. Yeah, of a hundred shares that I owned per every hundred that I owned, 10 of them I sold at 38 right at the direct listing. I want to say 40 of them I sold in the mid-20s, and the rest of it just got taken out at this price. So your dollar cost average to the, you know,

whatever high 30 is maybe 40 or something? Yeah, I don't know my exact. I mean, I sold some and I still own some. So, you know, I definitely got my beque wet from this acquisition. Oh, ho. But, no, but look, I think I probably sold, you know, more than half of them, you know. And that was a mistake. And, you know, one of my biggest learnings as an investor's has been to let your winners ride.

You know, my biggest mistake as an investor has not been the losers. It's all, it's been selling the winners, premer. surely. You did that with Uber as well, David. And I sold some Uber before, but I kept a lot of my Uber, maybe most of it or half of it, I think. Uber, Facebook. I mean, Facebook, you know, when they IPOed, it was worth $50 billion.

We all thought that was like unbelievable. I mean, because it was over a 50x return. What's the less than you sex? Does never sell anything if you can help it. I sold off my Facebook in 2014 and bought Amazon and Tesla. I think that you have to be able to sell for two reasons. Liquidity and moral obligation. Yeah, I mean, that's an exaggeration.

I mean, you can never, it's people need to be able to sell, but to the extent you can hold on, just don't sell everything. You know, always, you know, keep, you know, keep a piece of it. I mean, think about the people who were at Apple in the 80s or Microsoft in the 80s or Amazon in the 90s. A lot of those people got frustrated at holding the shares for so long. and I think keeping at least 20% of your shares forever, you know, could be amazing. There was somebody told me you had never sold a single share of a I don't know if that's a true story or not. I told you that.

You can't be leaked that information. Okay. Anyway, I didn't know that was a leak. Boop. Boop. More breaking news. Let's move to Alphabet. Let's move to Alphabet.

Must credit all-in podcast. Oh, my God. The St. Mayor may not be true without and his shares. You know what we should do is we should do a, we should put beeps in there, Nick. I was told B. Had never sold a share of B. And then we just let everybody react to it. This way nobody knows what we're talking about it.

I do know that B. Has not sold a single share. And it has only sold shares of BFund Capital Calls, which is an incredible statement to fortitude and vision. Incredible. Lord. Incredible. By the way,

by the way, it's not always worked out because he did the same with and those didn't go as well. Yeah. I mean, look, you have to diversify when you've got all your eggs in one basket and

one company. Obviously, you have to sell some shares. But, you know, one of the things I've just learned over the last 20 years is probably, you know,

people ask me what's your biggest regret or learning or whatever. It's just selling too early is like one of the biggest mistakes you can make. Look at PayPal. PayPal is now a $250 billion company. We sold it in 2002 for one. 1.5 billion. We thought that was a great deal at the time, and we sold it for less than 1% of what is worth today, and the product's basically the same.

You know, it's just compounding. So never sell is the lesson. Never sell. If it's a winner, write it. You can pair. Okay, hold on. Hold on. I'm going to put a final nail this off and then we're going to go to Alpha Fold.

There's a great quote by Warren Buffett, which is if you know what you're doing, the best thing you can do is be as concentrated as possible. Nobody ever got rich in their seventh best idea. And I think that that basically sums it up. But you have to be in a position to have the ability to have that kind of portfolio allocation. And I think that's hard. Freebird, explain alpha fold, please. Okay, let's explain, give me two minutes on, I'll explain proteins, and then the importance of proteins and then alpha fold.

So the numbers to remember are four, three, and 20. There are four nucleic acids that make up your DNA. We all learned this in high school biology. sets of three AC, T and G combinations define an amino acid. There are 20 amino acids. And a protein is a string of amino acids. So in your body in every cell, there are these organelles. They make proteins by reading the DNA, taking a copy of it,

and turning it into amino acid chains, and that's what we kind of call proteins. But what's interesting is when you make a chain of amino acid, so there's 20 of them, that you could put in each point in the chain, it doesn't come out as a long chain. What happens is those amino acids, the whole thing collapses, and it turns into a very specific shape. And the shape of that protein is what defines its function. So pretty much every biological function across all life is undertaken by proteins doing something.

Some proteins like hemoglobin in our red blood cells will have a very specific little pocket where oxygen molecules stick into the pocket, and then it moves the oxygen from your lungs to your cells. It's a pretty amazing protein to exist, and it specifically is shaped to do that exact function. There are other proteins that can, for example, rip apart other molecules, break a molecular bond. There are other proteins, for example,

that can take nitrogen out of the atmosphere and put it into plants cells that the plants can then use to grow. There's an incredible set of potential on the nanoscale of what you can do with proteins, and we see that in life and we're just shocked and awed and amazed by it every day. But in order to figure out how to create proteins that do specific things, you have to know how do those amino acids turn into the shape that the protein ultimately takes. And that's what's called protein folding. And so the hard thing is, and you know, why is this important? It's important because we can easily read DNA and therefore

we can figure out what amino acid sequence is being made to define that protein. But what we don't know really well is what is the shape of that protein and therefore how does it undertake the function that we see it taking in biology? And if you think about the reverse of this, the reverse of this, if you have a function you want to undertake in biology, you can design a protein to do that function for you. For example, bind to a specific point on a cancer cell or, you know, take carbon out of the atmosphere or pretty much anything else your mind can kind of imagine on the nanoscale proteins can be designed to do. the challenge is how do you write the code, which is the DNA, to make the protein that does that thing. Well, we don't know how the code turns into the shape. And that's what the folding problem is. So the folding problem, there's a data set.

And the data set is what's the three-dimensional shape of a protein? And then what's the DNA code that defines the amino acid sequence that makes that protein? And how do you figure out how to predict the shape of the protein from the amino acid sequence? It has been an impossibility. And again, if you think about this chain of amino acids, they each have little, you know, electrical spaces and the way that they bind to each other. It's very complicated. You can't just deterministically define it. You know, we don't have that level of understanding on a quantum scale.

So what Alpha Fold has done is they have now been able to predict from a sequence of amino acids what the protein shape will ultimately become. By learning from a database of hundreds of. thousands of structural protein shapes that have been defined through really, really, really difficult, you know, scanning microscopes and other techniques to really try and scan a protein on a microscopic scale. And then looking at the DNA sequence and figuring out, okay, what's the relationship? And the accuracy of their predictive model now is within the range of error of the microscopes that are being used to actually scan and measure those proteins.

So that's incredible because now theoretically you could come up with a design for a protein and you could actually build that protein by writing the amino acid sequence and that protein can do any number of things you want to do. And this has been a difficult problem that's been intractable by humanity and we've been challenged by it for decades. For this machine learning breakthrough to kind of be realized in literally less than three years, I mean, these guys were at a score of 40 last year and this year they're at like nearly 90.

which is incredible. And so now, you know, we can now predict what the shape will be from the DNA sequence. And this is going to unlock this ability. Everyone's now going to take their model if they license it or whatever they do with it. Or people are going to go learn using the same techniques that Deep Mind used. But it just means that it's possible. And then scientists will go away and they'll say, you know what, I want to do this particular thing on a microscopic scale. Let me design in three-dimensional space a protein to do that thing.

Okay, now let me go figure out how to make that process. by writing the DNA code, which is really easy if you can use this algorithm to solve that for you. And it is literally dollars and pennies to make proteins. We can write DNA on a computer. We can get printed DNA sent to us in 48 hours in a FedEx envelope for a DNA printing facility. We can put it in a microbe, and we can get that microbe to make the protein for us in a day. The lab costs, any high school biology class can do this now. So by being able to actually figure out what DNA to write, based on the objective function of what do we want the protein to do,

it's going to unlock this universe of things we can do in medicine, in environmental science. We can do things like break apart PET plastics. We can do things like fixing nitrogen from the atmosphere and getting rid of fertilizer plants. We can create all sorts of new food solutions, health solutions, environmental solutions. Any chance you can make a pizza that doesn't have carbohydrates, because that's what I'm thinking about here. Is there a way we can make a healthy pasta or pizza or something like that? but in all seriousness, what do you think the early wins will be out of this technology?

And is this a theoretical win that will benefit from in 20 years? Or is this a serious breakthrough that we're going to benefit from in the near term, like one to five years? Both are true. This is an incredibly important advancement in machine learning. But the reality is that Google will still have to spend a deep mind will have to spend a lot of time refining it. And then they have some really big ethical challenges ahead of it. how do you expose this technology to whom and under what conditions? And it's the same situation that Open AI has with GPT3, although a lot of people, I think, you know,

the scale of the computer science challenge maybe was a bigger win in GPT3 because it was a much more open space. And I think this is a much more specific sort of almost expert system in a way. But the downstream commercial implications of this are just enormous. And so just think about this. Like, this is where like you got to love companies like Google, the fact that they exist because from, you know, page rank in 1999 to CPC ads in 2003 and four, we have alpha fold in 2020. And that, to me, is just, that's just an end.

This is an argument against breaking up tech because only a tech company with this amount of resource knowledge can then go spend a billion dollars on deep mine over the past decade. Alphabet's burning $4 to $5 billion a year on their quote unquote other bets line. and people give them a lot of shit for it. But, I mean, you hit any one of these things, and it's $100 billion payday. I mean, look at YouTube. YouTube's easily $100 billion payday

on a billion dollar bet, billion six. Oh, no, that's a, that's a $250 to $500 billion company. Applied semantics. A lot of people miss this, but applied semantics was $100 million bet, and that's the entirety of AdSense initially. Android. Android.

Is this the first commercial application of DeepMind? Because until now, you know, they've had Alpha Zero. So there was a period of time. A lot of people, I don't know if, let me just think about this for a second. Because Alpha Zero, Alpha Zero was really good. I want to be careful about this, but I do think that. Not disclose. No, but I do think, I do think it was disclosed that they've used deep mind to improve ads

quality and to improve YouTube viewing. And as a result of that, you get the number of hours per day of the average user on YouTube to double or triple. Ad revenue goes by 3X. And I think in one quarter, Google was able to generate something like an incremental 15 billion annualized revenue from Deep Minds algorithm. And you know what that DeepMind actually did on YouTube? It sent everybody to the alt-ride and Info Wars and Ben Shapiro. Congratulations.

Be careful when you sent people's minds with artificial intelligence. Maybe you just argued to break them up. Yeah. But I understand Sacks that it's being used broadly across the products at Google. Now in a much more careful way is my understanding. Look, I'm a, I'm a big ally of Alpha Van. I'm a big fan. And I love, and I used to work there, and I'm very close to people there.

You know who used to be on the board and was the major backer of that company, DeepMind? Founders Fund. Founders Fund. Elon must, too. And he begged them to not sell to Google. I mean, $400 million exit was a steal for Google. With 40 or 50 scientists, I think he had. Yeah, absolute steel.

No, Elon, I mean, Elon has publicly said that he thinks DeepMind is like the greatest threat to, well, you think AI is a greatest threat to humanity. and of the people working on AI, Deep Mind is the furthest along and therefore most dangerous. He tried to stop. I mean, he told me straight up, like when he's been very public about this since,

that he said, I'll give you an unlimited amount of money to not sell to Google. He's friends with Larry Page too. But he said, don't sell. I want you to be independent. I want you keep working on this. But he said that when he saw the AI there

become aware that it was an AI, that was when he was like, wait a second. No, wait. Deep Mind is not. Alpha, alpha, folder, alpha zero is not self-aware. He felt it was becoming self-aware that it knew what it was. I don't know if that was just Elon, you know, just sort of, no, it's not, it's not self-aware.

So, I mean, I've watched, yeah, I mean, the amazing things that DeepMind has released prior to this were games, right? They had this, um, publicly. This chess AI called Alpha Zero, which rapidly became not just the best chest, um, it, not only beat every human in the world, it also beat every chess engine because computers became better at chess than humans a long time ago because of their sheer computational power. But Alpha Zero plays like a human, but with kind of that same computational power. And so that created a whole revolution in chess engines.

And then they also did that with a game called Go. They created AlphaGo. And basically, every single game that you can think of, Deep Minds created an alpha, whatever, Alpha version that destroys both humans and computers. But this is the first thing they've publicly announced that that seems like it will be available to others eventually that could have tremendous social impact. Imagine the government trying to understand this. Can you imagine them being brought before like senators and Congress?

These are these are the new weapons of mass destruction. I mean, let's be honest. Like if if somebody else had had Alpha Fold and probably some, somebody does and just hasn't said. You know, I mean, Google actually values transparency. So that's the only reason why we know. Imagine what they could do, as Friedberg said. It's like, you know, the opposite of this is basically to design a specific protein that basically, you know, destroys organs or...

There are proteins called prions, which are the scariest thing known in biology, in my opinion. A prion is a protein that it actually finds similar proteins. And based on its shape, it gets those proteins to change. and it becomes like a virus. And prions actually are, there's an extremely sad series of diseases that are related to prions where your body expresses a protein in the wrong way. And then that protein itself gets other proteins to change and creates copies of itself and it spreads.

It is a fascinatingly scary biological phenomenon. But, you know, there are extremely scary things you can do with the designer protein capabilities in the wrong hands. To listen to the rest of the podcast, search for All In with Chmoth, Jason, Sacks, and Friedberg, available across all major podcasting platforms.