Skip to content
All-In PodcastAll-In Podcast

Elon Musk: Grokipedia, OpenAI betrayal, and free speech

Grokipedia trains Grok to break arguments into axioms and cross-check sources; Community Notes already showed Wikipedia bias on political and tribal topics.

Jason CalacanishostChamath PalihapitiyahostDavid FriedberghostElon Muskguest
Oct 31, 20251h 33mWatch on YouTube ↗

EVERY SPOKEN WORD

  1. 0:00 – 3:10

    Disgraziad Corner: The most disgraceful things of the week!

    1. JC

      Let's get started. You know, we wanted to try something new this week. Every week, uh, you know, I get a little upset, things perturb me, Sacks, and, uh, when it does, I just yell and scream, "Disgraciad!" And so I bought the domain name disgraciad.com for no reason other than my own amusement. But you know what? I- I'm not alone in my absolute disgust at what's going on in the world. So this week, we're gonna bring out a new feature here on the All-In Podcast, Disgraciad Corner. (instrumental music plays)

    2. CP

      (laughs)

    3. JC

      (laughs)

    4. DS

      Disgraciad.

    5. NA

      He was the best guy around.

    6. DF

      What about the people he murdered?

    7. NA

      What murder?

    8. DF

      You can act like a man. That's what I'll do.

    9. NA

      (laughs)

    10. DF

      He's just killed all the (beep) mayors. He insulted them a little bit. I'm smart and I want to spit. What's the matter with you? Your hair was in the toilet water. Disgusting. I oughta suffocate you, you little (beep) .

    11. JC

      (laughs)

    12. DF

      It's a (beep) disgrace.

    13. JC

      Wow.

    14. DS

      Disgraciad.

    15. JC

      Disgraciad.

    16. CP

      This is fantastic. Wow. What a-

    17. JC

      This is our new feature. Chamath, you look like you're ready to go. Why don't you tell you- tell everybody who gets your disgraciad this week?

    18. CP

      Wait, we all have to come with a disgraciad? (laughs)

    19. JC

      You really (laughs) person- you, sorry, you missed the memo.

    20. CP

      (laughs)

    21. (laughs)

    22. JC

      All right, fine. Enough.

    23. CP

      I got one, I got one.

    24. JC

      Okay, all right. Just calm down.

    25. CP

      My disgraciad corner goes to Jason Calacanis-

    26. (laughs)

    27. JC

      Oh, here we go. (laughs) Come on, man. You can't-

    28. CP

      ... and Pete Buttigieg where they, in the first 30 seconds of the interview-

    29. JC

      (laughs)

    30. CP

      ... compared virtue signaling points about how each one worked at various moments at Amnesty International.

  2. 3:10 – 11:35

    Elon on X's new algorithm, why there has been so much Sydney Sweeney content lately

    1. NA

    2. JC

      What's going on with the algorithm? I'm getting Sydney Sweeney's dress all day and last week-

    3. DS

      (laughs)

    4. JC

      ... Sacks-

    5. EM

      Well, maybe you should stop favoriting it.

    6. CP

      (laughs)

    7. JC

      I couldn't. I favorited it 15 times.

    8. DS

      (laughs)

    9. JC

      And then Sacks, poor Sacks got... You got invited to SlackCon for two weeks straight on the algorithm.

    10. EM

      What's, uh...

    11. CP

      No, I say the algorithm has become-

    12. EM

      What is SlackCon?

    13. CP

      If you, if you demonstrate-

    14. JC

      (laughs)

    15. DS

      Sacks will explain. (laughs)

    16. EM

      You can't even tell if that's a joke or a real thing.

    17. JC

      It's a real thing.

    18. CP

      No, it's real.

    19. DS

      It's a real thing.

    20. JC

      It's San Francisco.

    21. CP

      It's all too real.

    22. DS

      Oh, it, it's actually (beep) real? Wait.

    23. CP

      Yeah.

    24. DS

      SlackCon is a real... For real?

    25. CP

      What I've noticed... Yeah, if you, if you demonstrate interest-

    26. DS

      (laughs)

    27. CP

      ... in anything on X now, if you click on it, god forbid you like something.

    28. DS

      (laughs)

    29. CP

      Man.

    30. JC

      Yeah.

  3. 11:35 – 24:52

    Creating Grokipedia: Wikipedia's failures, the future of information on the internet, confirmation bias

    1. EM

      The recent update to Grok is actually, I think, very good, um, at piercing through propaganda. Um, so, um, and we, we, we used that latest version of Grok to create Grokipedia, um, which I think is, um, uh, much, uh, more... It- it- it's, it's not just, I think, more, um, neutral, um, th- than- and, and more accurate than, than Wikipedia, but it actually has a lot more information than a Wikipedia page.

    2. JC

      Did you seed it with Wikipedia? Or actually, take a step back. How did you guys... How did you do this?

    3. EM

      Um, well, we used AI. (laughs)

    4. JC

      (laughs)

    5. CP

      But meaning, like, totally unsupervised, just complete training run on its own, totally synthetic data, no, no seeded set, nothing?

    6. EM

      Um, well, it- it was only just, uh, recently possible for us to do this. Um, so we've, um, we- we- we finished training on a maximally truth-seeking... w- w- well, a version of Grok that is good at cogent analysis, so breaking down, um, uh, any- any given argument into its axiomatic elements, um, assessing whether those a- axioms are, um... n- you know, the- the basic test for cogency, the axioms are likely to be true, they're not, um, contradictory, uh, that, um, the conclusion naturally... that- that- the conclusion most likely follows from those axioms. Um, so- so we're- we're just trained Grok on a lot of critical thinking, um, so it just got really good at critical thinking, um, which was qu- quite hard. And then we ta- took that version of Grok and said, "Okay, cycle through the- the million most, uh, popular, uh, articles on Wikipedia and- and add, modify and delete." So, um, that means, um, research the rest of the internet, um, whatever's publicly available, um, and correct, uh, what's... correct Wikipedia articles and fix mistakes, um, but also add a lot more context. Um, so- so- sometimes, really, the nature of the propaganda is that, um, you know, facts are stated that are- that are technically true, but are not repre- do not properly represent a picture of the individual or event.

    7. JC

      This is critical because when you have a bio, as you do (laughs) , actually, we all do, on Wikipedia, over time, it's just the people you fired or you- you beat in business or have an ax to grind, so it just slowly becomes, like, the place where everybody, you know, kind of who hates you then puts their information. I looked at mine, it was so much more representative and it was five times longer, six times longer.

    8. EM

      Yeah.

    9. JC

      And the... what it gave way to, uh, was much more accurate.

    10. EM

      Yeah.

    11. JC

      Much more accurate. And this opportunity was sitting here I think for a long time.

    12. EM

      Yeah.

    13. JC

      Uh, and it's just great that you got to it because they- they don't update my page but, you know, I don't know, twice a month with, you know, and then who is this secret cabal that's 50 people who are anonymous who decide what's get, gets put on it? It was a much better, much more updated page-

    14. EM

      Yeah.

    15. JC

      ... in version one.

    16. EM

      Uh, yes, this is version 0.1 as we've put it... as we show at the top. So, um, I- I do think actually by the time we get to version one, 1.0, it'll be 10 times better. But even at this early stage, um, as you, as you just mentioned, it's- it's not just that it's correcting errors, but, um, it- it is creating a- a more accurate and realistic and fleshed out, uh-

    17. JC

      Mm-hmm.

    18. EM

      ... description of- of people and events.

    19. JC

      Elon, do you think that-

    20. EM

      And- and- and- and subject matters. Like, you can look at articles on- on physics in Grokipedia, they're- they're much better than Wikipedia by far.

    21. JC

      That's what I was gonna ask you, is like, do you think that you can take this corpus of pages now and get Google to de-boost Wikipedia or boost Grokipedia in traditional search? Because a lot of people still find this and they believe that it's authoritative because it comes up number one.

    22. EM

      Yeah.

    23. JC

      Right? So how do we, how do we do that? How do you flip Google?

    24. EM

      Um, yeah. So it- it really can... if- if people share a lot of... if- if- if Grokipedia, uh, is used elsewhere, like, if people cite it on their websites, uh, or post about it on social media, um, or when they do a search where Grokipedia shows up, they click on Grokipedia, it will naturally, uh, uh, you know, rise in- in Google's, uh, you know, rankings. Um, I did- I did send (laughs) ... I did tag Sundar because, you know, even- even sort of the day after launch, if you typed in Grokipedia, Google would just say, "Did you mean Wikipedia?"

    25. JC

      Wikipedia, yeah.

    26. (laughs)

    27. EM

      And it wouldn't even bring Grokipedia up at all.

    28. JC

      Yeah. That's true.

    29. EM

      So- so now-

    30. JC

      How's the u- how's the usage been? Have you seen good growth since it launched?

  4. 24:52 – 42:49

    Three years of X: Looking back on the Twitter acquisition and how it changed free speech on the internet

    1. CP

      I think we originally had the idea to have you back on the pod because it was the three-year anniversary of the Twitter acquisition. So-

    2. EM

      Okay. (laughs)

    3. CP

      ... I, I just wanted to kinda reminisce a little bit.

    4. EM

      Yeah, sure.

    5. CP

      And I remember... Yeah. I mean, I remember-

    6. EM

      Where's that sink?

    7. CP

      (laughs) Where's that sink? Well, yeah.

    8. JC

      (laughs)

    9. CP

      So Elon was staying at my house. We had talked the week before and he told me the deal was gonna close, and so I was like, "Hey, do you need a place to stay?" And he took me up on it. And the day before he went to the Twitter office, there was a request made to m- to my staff, "Do you happen to have an extra sink?" And they did not, but they were able to, uh-

    10. JC

      (laughs)

    11. EM

      Yeah. Who, who has an extra sink, really?

    12. CP

      (laughs)

    13. JC

      (laughs)

    14. EM

      (laughs)

    15. CP

      But they were able to f- to locate one at a nearby hardware store, and I think they paid extra to get it out of the window or something.

    16. EM

      (laughs) Well, I, I think the, the store was confused because, um, my security team was asking for, uh, any kind of sink. And, and, and like-

    17. JC

      (laughs)

    18. EM

      Like, normally people wouldn't ask for any kind of sink. (laughs)

    19. JC

      (laughs)

    20. CP

      (laughs)

    21. EM

      You need a sink that fits in your bathroom or connects with certain kinda plumbing. So they're like trying to ask me, he's like, "Well, what kind of faucets do you want?" That's... No, no, I just wanted a sink. (laughs)

    22. JC

      Yeah, they think it's a mental person calling. (laughs)

    23. EM

      They're like, the, the store was confused that we just wanted a sink.

    24. CP

      Yes.

    25. EM

      And, and didn't, and didn't care what, uh, what the sink connected to. (laughs) I mean-

    26. CP

      That was, that was a really-

    27. EM

      They were just like, they, they were like almost not letting us buy the sink because, because they, they thought maybe we would buy the wrong sink, you know? Um, it's just rare that somebody wants a sink for sake's sake.

    28. JC

      (laughs)

    29. CP

      (laughs) For meme purposes.

    30. JC

      One of my favorite memories was Elon said, "Hey, you know, swing by, check it out." I said, "Okay, I'll come by." And I drive up there and I'm looking where to park the car, and I realize there's just parking spaces around the entire bus- building. And I'm like, "Okay, this can't be, like, legal parking," but I park and it's-

  5. 42:49 – 47:40

    Tesla vote on Elon's compensation, would he leave Tesla if it doesn't pass?

    1. EM

    2. CP

      Can I ask you a question just to maybe move to a different topic? If you came and did this next week, we will be past the Tesla board vote. We talked about it last week, and we talked about how crazy ISS and Glass Lewis is, and-

    3. EM

      Right.

    4. CP

      ... we used this one insane example where like, Ira Anropase didn't get the recommendation from ISS and Glass Lewis because he didn't meet the gender requirements, but then Kathleen also didn't.

    5. EM

      Yeah, no, no, it, it, it doesn't make any sense.

    6. CP

      Can you... So-

    7. EM

      Um-

    8. CP

      ... the, the board vote is on the 6th so-

    9. EM

      Kath- Kath- Kathleen was an African American woman.

    10. CP

      Yeah.

    11. EM

      (laughs) Yeah. She, she was... They recommended against her, but then also recommended against, uh, R. Anropase, um, on, on the grounds that he was insufficiently diverse. So I'm like, this, like these things don't make any sense.

    12. CP

      Yeah.

    13. EM

      So, I, I do think we've got a fundamental issue with corporate governance, um, in publicly traded companies, where you've got about half of the stock market, uh, is controlled by passive index funds, um, and, uh, most of them out- most of them outsource their decision, uh, to, uh, advisory firms, and par- particularly Glass Lewis and, uh, ISS. I call them Corporate ISIS. Um-

    14. CP

      (laughs) .

    15. EM

      You know, so it's, it's, it's... All they do is basically... They're just, they're just terrorists. Um, so, um, so... And, and, and they have, they own no stock in any of these companies.

    16. CP

      Um, right.

    17. EM

      So I, I think that there's, there's a fundamental breakdown of fiduciary responsibility here, uh, where really, um, you know, a- any company that's managing, um, uh, even though they're passively managing, you know, index funds or whatever, they, they do, at the end of the day, have a fiduciary duty to, uh, vote, uh, you know, along the lines of what would maximize the, the shareholder returns, because people are counting on them. Like, people, uh, you know, have, say... You know, so has- have all their savings in, say, a 401 (k) or something like that, um, and they're, they're counting on, um, the index funds to vote, uh, do company votes in the direction that would, uh, ensure that their retirement savings, uh, do as well as possible. But the problem is if that is then outsourced to ISS and Glass Lewis, which have been infiltrated by far-left activists, um... Because you know where, you know where far left- a- a... You know, you know, you know where, where basically political activists go. They go where the, where the power is.

    18. CP

      Mm-hmm.

    19. EM

      Um, and so effectively, uh, Glass Lewis and ISS, uh, con- uh, control the vote of half the stock market. Now, now if, if you're a political activist, you know what a great place would be to go work? ISS and Glass Lewis, and they do. So, um, so my concern for the future, um... 'Cause there's a... The, the, you know, the Tesla, um, thing is, is it's called sort of compensation, but really it's not about compensation. It's not like I'm, like I'm gonna go out and buy, you know, a yacht with it or something. It's just that I, I, I do... I... In order... If I'm going to, uh, build up, uh, Optimus and, and, you know, have all these robots out there, I need to make sure we do not have a Terminator scenario, and I... And that I can make... You know, maximize the safety of the robots. Um, and, and, and, um... But I, but I, I, I feel like I, I need to have something like a 25% vote, um, which is enough of a vote to have a strong influence, uh, but not so much of a vote that I can't be fired if I go insane. Um, (laughs) so it's, it's kinda... But, but my concern would be, you know, creating this army of robots and then, and then being fired for political reasons, um, because of, because of ISS and Glass Lewis, uh, uh, you know, declined to... I- i- ISS and Glass Lewis fire me effectively or, or the, the activists at those firms fire me, um, even though I've done everything right.

    20. CP

      Yeah.

    21. EM

      That's my concern.

    22. CP

      Yeah.

    23. EM

      And then I, and then, then, then you've got... And then I, and then I cannot ensure the s- the safety of the robots.

    24. JC

      If you don't get that vote, if it doesn't go your way, it looks like it's going to, would you leave? I mean, is that even in the cards? I heard they were... The board was very concerned about that.

    25. EM

      Uh... Let's just say I'm not gonna build a robot army, um, if I, if I can be easily kicked out by activist investors.

    26. JC

      Yeah.

    27. EM

      No way.

    28. JC

      All right. Yeah. Makes sense. I mean-

    29. EM

      Yeah.

    30. JC

      And who is capable of running the four or five major product lines at Tesla? I mean, this is the, the madness of it. It's a very complex business. People don't understand what's under the hood there. It's not just a car company. You got batteries, uh, you got trucks, you got the self-driving group, and this is a very complex buil- business-

  6. 47:40 – 56:24

    OpenAI lawsuit, for-profit conversion, how much Elon should own, OpenAI's great irony

    1. JC

    2. CP

      By the way, if we take a full turn around corporate governance corner, also this week, what was interesting about the OpenAI restructuring was I read the letter, and-

    3. EM

      Yeah.

    4. CP

      ... your lawsuit was excluded from the allowances of the California Attorney General basically saying this thing can go through, which means that your lawsuit is still out there, right? And I think it's gonna go to a jury trial.

    5. EM

      Yes.

    6. CP

      So there, that corporate governance thing is still very much in question. Do you have any thoughts on that?

    7. EM

      Um, yes, I believe that will go to a jury trial in February or March, um, and, and then we'll, we'll see what the re- what the results are there. But, um, I, I... There's, there's an o- an... Like a, a mountain of evidence, um, that, that shows that OpenAI was created as a, uh, an open source nonprofit. It's, it's literally... That's the, the exact description in the incorporation documents. Um, and in fact, the incorporation documents explicitly say that no officer, uh, or founding member will be... Will, will benefit financially from OpenAI.

    8. CP

      Right.

    9. JC

      Hmm.

    10. EM

      And that they've completely violated that. And moreover, you can... That then- we're- you can just use the Wayback Machine and look at the website of OpenAI. Again, open source, non-profit, open source, non-profit, the whole way, until, you know, it looked, it looked like, "Wow, this is, uh, there's a lot of money to be gained here." And then suddenly it starts changing. Um, and they try to change the definition of OpenAI to mean open to everyone instead of open source, even though it always meant open source.

    11. JC

      Hmm.

    12. EM

      I came up with the name. (laughs)

    13. JC

      Yeah. (laughs) Uh-

    14. EM

      That's how I know.

    15. CP

      (laughs)

    16. JC

      So, uh, if they open sourced it, uh, or they gave you... I mean, you don't need the money, but if they gave you the percentage ownership in it that you would be rightfully, uh, which 50 million for a startup would be half, (laughs) at least-

    17. EM

      Yeah.

    18. JC

      ... but they must have made an overture towards you and said, "Hey, can we just give you 10% of this thing and give us your blessing?" Like prov-

    19. EM

      Um-

    20. JC

      You obviously have a different goal here, yeah?

    21. EM

      Yeah, um, e- I mean, essentially, since I came up with the idea for the company, named it, um, pro- provided the A, B and C rounds of funding, uh, recruited the, uh, critical personnel, uh, I told them everything I know. Um, you know, if that had been a, a commercial corporation, I'd probably own half the company. So, um, but... And, and, and I could, I could have chosen to do that. That, that's... I... If I... It was totally at my discretion. I could have done that. Uh, but I created it as a nonprofit for the world, an open source nonprofit for the world.

    22. CP

      Do you think the right thing to do is to take those models and just open source them today? If you could effect that change, is that the right thing to do?

    23. EM

      Uh, yeah, I think, I think, uh, that, that, that is what the co- what it was created to do, so it should. I mean, the b- the best open source models right now, actually, ironically, uh, because fate, fate seems to be an irony maximizer-

    24. CP

      Mm-hmm.

    25. EM

      ... um, uh, the best open source models are generally from China.

    26. JC

      Yeah.

    27. EM

      Like... (laughs) That's bizarre. And, and, and then the... I think the second-best, uh, uh, is... or one is... or may- maybe it's better than second best, uh, but like the, uh, the Grok 2.5, um, open source model is actually very good. Um, and I think what would probably be... A- a- and we'll continue to open source our models, but, you know, but whereas, like, try using any of the, the, the recent, um, so-called, the OpenAI open source models, they're out, they don't work. They basically... They open sourced a broken, non-working version of, of their models, uh, as a fig leaf. I mean, do you know anyone who's running open, OpenAI's open source models? Exactly.

    28. JC

      Yeah, nobody. We've had a big debate about jobs here. Obviously, there's gonna be job displacement. You and I have talked about it for decades. Uh, what's your take on the pace of it? Because obviously you're building self-driving software, you're building Optimus.

    29. EM

      Yeah.

    30. JC

      And we're seeing Amazon take some steps here where they're like, "Yeah, we're probably not gonna hire these positions in the future." And, you know, uh, maybe they're getting rid of people now because they were bloated, but maybe some of it's AI, you know, it's, it's all debatable. What do you think the timeline is and what do you think, as a society, we're gonna need to do to mitigate it if it goes too fast?

  7. 56:24 – 1:09:34

    AI power efficiency, Robotaxis, future of self-driving

    1. DS

      that possible? Yeah.

    2. CP

      In, in the earnings call, you said something that was insane, and then I think the math actually nets out. But you said we could connect all the Teslas and allow them in downtime to actually offer up inference.

    3. EM

      Yeah.

    4. CP

      And you can string them all together. (laughs) I think the math is like, it could actually be like 100 gigawatts.

    5. DS

      Oh, is that right?

    6. CP

      Is that right? Or did you do-

    7. EM

      If, if ultimately th- there's a Tesla fleet that is, um, uh, 100 million vehicles, uh, which I think we probably will get to at some point, 100 million vehicle fleet, um, and, uh, they have s- you know, mostly state-of-the-art, uh, inference computers in them, uh, that, that each say are, uh, a, a kilowatt of inference compute. Um, and they have built in, um, power and cooling, um, and-

    8. DS

      Yeah. Yeah.

    9. EM

      ... you know, connected to the wifi with it.

    10. CP

      That's the key. Yeah, exactly.

    11. EM

      Um, yeah. Exactly. And, and, and, and, uh, that the en- you'd have a 100 gigawatts of inference compute.

    12. DS

      Elon, do you think that the architecture, like there was an attention-free model that came out in the last week. There's been all of these papers, all of these new models that have been shown to reduce power per token of output by many, many, many orders of magnitude. Like not just an order of magnitude, but like maybe three or four. Like what's your view in all the work you've been doing on where-

    13. EM

      Yeah.

    14. DS

      ... we're headed in terms of power, um, uh, per unit of compute or per token of output?

    15. EM

      Well, it, it, we, we have a, a clear example of, uh, efficient, power-efficient compute, which is the human brain. Um, so, um, our brains use about 20 watts, um, of power, but, and of that, only about 10 watts is higher brain function. Most of it's, you know, like half of it is just housekeeping functions, uh, you know, keeping your heart going and breathing and that kind of thing. Um, so, so you've got maybe 10 watts of, uh, higher brain function in a human. Um, and we've managed to build civilization with 10 watts of, uh, of a biological computer. Um, and that biological computer has like a 20-year, you know, boot sequence.

    16. DS

      Yeah. (laughs)

    17. EM

      Uh, so it's pretty, pretty, uh... (laughs) but, but, but it's very power efficient. So, uh, given that, uh, humans are capable of inventing, um, you know, general relativity and quantum mechanics and, uh, or discovering ge- general opportunity, like, like inventing aircraft, lasers, the internet, and discovering physics with a, with a 10-watt, uh, meat computer essentially. Um, uh, then, um, there's clearly a m- a, a massive opportunity for improving the, uh, efficiency of AI compute. Um, it's 'cause it's, it's, it's currently many orders of magnitude away from that. Um, and, and it's still the case that, um, 100 megawatt, uh, or even, you know, a gigawatt, uh, AI supercomputer at this point can't do everything that, uh, a human can do. Uh, it, it will be able to, uh, but it can't yet. Um, so, but, but we, we, we ha- like say we've got this obvious case of, um, uh, human brains being very power efficient and achieving and, and building civilization with, with a, with a, you know, with, with 10 watts of compute. Um, and, and, and a very slow... A- a- and our, uh, our bandwidth is very low, so the, the, the speed at which we communicate information to each other is extremely low. You know, we're, we're not communicating at a terabyte. We're communicating more at like 10 bits per second. Um, so, um...

    18. CP

      Do you think that there's a-

    19. EM

      A- and that should naturally lead you to the conclusion that there's massive, uh, opportunity for be- being more power efficient with, with AI. And, and at Tesla and at xAI, we're both, we, we, we continue to see massive improvements in, in inference computer efficiency. Um, so, um, yeah.

    20. CP

      You think that there's a moment where you would justify stopping all the traditional cars and just going completely all in on Cybercab if you felt like the learning was good enough and that the, the system was safe enough? Is th- is there ever a moment like that or do you think you'll always kind of dual track and always do both?

    21. EM

      I mean, all of the cars we make right now, um, are capable of being a robotaxi. So there's a little confusion of the terminology 'cause, um...... the, the, the, our cars look normal, you know, like Model 3 or Model Y looks ... It's a good-looking car, but it looks, looks normal. Um, but it has an advanced AI computer and advanced AI software and cameras, and we didn't want the cameras to stick out, so we, you know, so that ... We didn't want them to be ugly or stick out. So, so, you know, we put them ... they're sort of in unobtrusive locations. You know, the forward-looking camera, cameras are in front of the rear-view mirror. Um, the side-view mirrors are in the side repeaters. Or the, the side-view cameras are on the side repeaters. Um, the, the, the rear camera is, you know, just in the, you know, uh, above the license plate, actually typically where the review camera is in a car. Um, and, um, you know, and, and, and the, the diagonal forward ones are in the B pillars. Like, if you look closely, you can see all the cameras, but, but you have to look closely s- ... We just didn't want them to be ... to stick out like, you know, warts or something. Um, but, but actually, all the cars we make, um, are hyper-intelligent, um, and have the cameras in the right places. They just look normal. Um, and, um, so, so all of the cars we make are capable of unsupervised f- full autonomy. Um, now, we, we have a dedicated product, which is the Cybercab, um, which has no, no steering wheel or pedals, um, which are obviously prestigial in a autonomous world. Uh, and we start production, uh, of the Cybercab in Q2 next year, and we'll scale that up to, to quite high volume. I think ultimately we'll make millions of, uh, Cybercabs per year. Um, but, but it, but it is important to emphasize that all of our cars are capable of being robotic taxis. Um-

    22. JC

      The Cybercab is gorgeous. I told you I'd buy two of those if you put a steering wheel in them, and there is a big movement online-

    23. EM

      We're not putting a steering wheel.

    24. JC

      People are begging for it. Why not? Why not let us buy a couple, uh, you know, ugh, you know, just the first ones off the line and drive them? I mean, it's ... They look great. It's like the perfect-

    25. EM

      Well-

    26. JC

      ... model. You always had a vision for a Model 2, right? Like, isn't it like the perfect Model 2 in addition to being a, a Cybercab?

    27. EM

      Look, the reality is p- people may think they wanna drive their car, but the reality is that they don't. Um, uh, uh, uh, how many times have you been to, say, an Uber or Lyft and, and you said, "Do you know what? I wish I could take over from the driver."

    28. JC

      (laughs)

    29. EM

      (laughs) And, and, and, and I wish I could get off my phone and, and take over from the Uber driver and, uh, and, and drive to my destination." How many times have you thought to yours- ... thought that to yourself?

    30. JC

      No, I ... It's quite the opposite.

  8. 1:09:34 – 1:17:13

    Bill Gates flips on climate change, solar, energy production

    1. JC

      ask if you saw the Bill Gates memo that he put out?

    2. EM

      (laughs)

    3. JC

      A lot of people are talking about this memo.

    4. (laughs)

    5. EM

      Yeah.

    6. JC

      Like, you know, did-

    7. EM

      Yeah.

    8. JC

      I- I- I just want to say-

    9. EM

      I just wanna say, Billy G is not my lover.

    10. JC

      (laughs)

    11. (laughs)

    12. Oh, man.

    13. EM

      (laughs)

    14. JC

      Like, did- did- did climate change-

    15. DS

      Oh, man.

    16. JC

      ... become woke?

    17. EM

      (laughs)

    18. JC

      Did it become, like, woke, and is it over being woke?

    19. EM

      (laughs)

    20. JC

      Like, you know, like, what happened and what's, what- what happened with-

    21. DS

      Oh.

    22. JC

      ... Billy G? I mean ...

    23. EM

      (laughs)

    24. JC

      You know.

    25. Short and Tesla. (laughs)

    26. DS

      Great question. Great question.

    27. JC

      Yeah.

    28. EM

      Uh, y- y- you know, you- you'd think that someone like Bill Gates, who, uh, clearly started a tech- ... You know, started- started a technology company that's one of the- the biggest companies in the world, Microsoft. Um, being, uh, y- y- you'd think he'd be really quite, um, you know, strong in the sciences. Um, but actually, my, at least direct conversations with him have, um ... He- he- he is, he is not strong in the sciences. Like, the ca- the- he d- ... Yeah. This is s- really sh- surprising. You know, like, he- he came to visit me at the Tesla gigafactory in, in Austin and was telling me that it's impossible to have a long ranged, uh, semi truck. Um, and I was like, "Well, but we literally have them. Um, and you can drive them, and Pepsi is literally using them right now, and you can drive them yourself or send someone." Obviously Bill Gates is not gonna drive it himself, but, "You can send, uh, a trusted a- uh, person to drive, like, the truck and verify that it can do the things that we say it's doing." And he's like, "No, no, it doesn't work, doesn't work." And I'm like, "Um, okay."

    29. DS

      (laughs)

    30. EM

      I'm like kind of stuck there.

Episode duration: 1:33:35

Install uListen for AI-powered chat & search across the full episode — Get Full Transcript

Transcript of episode j6_VfR-CyuM

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome