WEBVTT 1 00:00:00.280 --> 00:00:04.480 I'm certainly saying that that is their policy, and I think you should take 2 00:00:04.480 --> 00:00:08.039 their words seriously. Like, I work seven days a week. But but 3 00:00:08.160 --> 00:00:13.359 I I'm not expecting others to do that. Offer me money, offer me 4 00:00:13.480 --> 00:00:19.280 power, I don't care. Yeah, this this is actually where, UM, 5 00:00:19.600 --> 00:00:22.879 I don't actually have an office here. I either walk the floor or 6 00:00:22.920 --> 00:00:27.239 I hold meetings in this room. Okay, And this room allows me to 7 00:00:27.359 --> 00:00:32.000 see what's going on with production while I'm holding meetings. Is there one particular 8 00:00:32.039 --> 00:00:35.119 production that you focus on when you're here or that you sort of your eyes 9 00:00:35.159 --> 00:00:39.039 always go to, or you you actually walk over to. Um, Well, 10 00:00:39.039 --> 00:00:42.240 this is the end of line that we're seeing here. So if if 11 00:00:42.439 --> 00:00:47.240 cars aren't moving, um, that means there's some blockage upstream. And then 12 00:00:47.240 --> 00:00:50.960 I can go and see what that blockage is. UM. So I think 13 00:00:50.960 --> 00:00:55.159 we might be on a break right now. But this if this, this 14 00:00:55.719 --> 00:00:59.240 is if this, if a vehicle end of line is not moving, that 15 00:00:59.280 --> 00:01:02.359 means there's something up stream that's wrong. Got it. Well, it's been 16 00:01:02.359 --> 00:01:04.120 amazing just to walk around the facility a bit and get a sense for the 17 00:01:04.159 --> 00:01:07.519 size. All right, you just finished your annual meeting. You took a 18 00:01:07.519 --> 00:01:12.120 lot of questions from a very eager audience. I feel like most of it 19 00:01:12.120 --> 00:01:15.599 you've shared before, but there was some news and I want to start there. 20 00:01:17.359 --> 00:01:19.879 Advertising. You got a big applause. You didn't sound like you were 21 00:01:19.879 --> 00:01:23.200 going to advertise that much, but you did at least seem to open the 22 00:01:23.239 --> 00:01:27.439 way to saying, Okay, we might consider doing some advertising. Why. 23 00:01:29.599 --> 00:01:33.439 Well, I mean, I've believen listened to listening to shareholders, and I 24 00:01:33.439 --> 00:01:37.439 actually was surprised by the level of enthusiasm for advertising, since we are not 25 00:01:37.480 --> 00:01:42.280 historically done that. But there perhaps is some good logic to it in that 26 00:01:44.640 --> 00:01:49.120 if we're simply saying that information by I say that the Tesla Twitter account on 27 00:01:49.120 --> 00:01:53.120 my Twitter account, we're somewhat preaching to the converted and not reaching people that 28 00:01:55.000 --> 00:01:59.040 are not already convinced. Essentially, so I think they're probably have a good 29 00:01:59.040 --> 00:02:00.760 point. Well, I mean, I think it's worth a try and we'll 30 00:02:00.799 --> 00:02:05.120 see, we'll see how factor Do you have any idea how you'd like to 31 00:02:05.120 --> 00:02:08.240 try it or what that would involve. Um Well, I mean I have 32 00:02:08.240 --> 00:02:15.360 some general thoughts about advertising. Um that that you know, if advertising is 33 00:02:15.680 --> 00:02:23.439 informative, um and and entertaining, then it is it can start to approach 34 00:02:23.520 --> 00:02:30.479 content. UM. So I think sometimes advertising is perhaps a non informative or 35 00:02:30.560 --> 00:02:36.199 perhaps in some cases a bit misleading. Um. And in fact, we've 36 00:02:36.240 --> 00:02:40.599 lost some advertisers on Twitter because community notes applies to advertising too, and so 37 00:02:40.719 --> 00:02:47.439 if somebody advertises something that isn't, um perhaps a bit inaccurate, then it 38 00:02:47.479 --> 00:02:52.759 gets community noted, and then they get mad and stop advertising. But we're 39 00:02:53.360 --> 00:02:57.759 we're care enough about the truth that we're willing to give out give up advertising 40 00:02:57.759 --> 00:03:02.840 dollars on Twitter, even if in order to having the worst, the least 41 00:03:02.879 --> 00:03:07.360 inaccurate, right, And I want to talk about that, but I want 42 00:03:07.360 --> 00:03:09.520 to obviously spend some time on Tesla again. Back to adds though, I 43 00:03:09.560 --> 00:03:13.680 mean, am I going to start to see test lads during the NFL broadcast 44 00:03:13.759 --> 00:03:17.840 on a given Sunday Monday Thursday? The NFL's always on, well bar in 45 00:03:17.840 --> 00:03:20.960 mind. I mean I only just agree to it, so it's not like 46 00:03:20.960 --> 00:03:23.080 I have a pulle formal strategy. Okay, Well, how much time does 47 00:03:23.120 --> 00:03:25.639 go into when you think about these things? I mean, was it something 48 00:03:25.680 --> 00:03:30.000 you decided to do up on stage or at least did you know coming in? 49 00:03:30.319 --> 00:03:32.919 I may you know this comes up that I may go there. No, 50 00:03:34.080 --> 00:03:38.639 I mean the questions are not scripted in any way, right, So 51 00:03:38.639 --> 00:03:40.439 so in that moment, you just said, all right, I'll do some 52 00:03:40.479 --> 00:03:45.240 ads. That's right. So we're only like twenty minutes later and I'm asking 53 00:03:45.240 --> 00:03:47.759 you detailed questions. Yeah, have any answers? I guess I'm not sure 54 00:03:47.759 --> 00:03:50.800 what the most spectrum thing is, except that, like I said, I 55 00:03:50.800 --> 00:03:57.080 believe that advertising should be informative about a product. It should be ideally esthetically 56 00:03:57.159 --> 00:04:00.639 pleasing, should be you know, beautiful to some autistic element to it, 57 00:04:00.360 --> 00:04:05.479 and it should be something that you don't regret watching after it's done. And 58 00:04:05.599 --> 00:04:12.039 I think if advertising fits those criteria, that it starts to approach content, 59 00:04:12.199 --> 00:04:16.199 and you wan't advertising that is as close to content as possible such that you 60 00:04:16.240 --> 00:04:19.519 don't regret the time you spent watching it. But you did say something as 61 00:04:19.519 --> 00:04:23.319 well, which is that you know you've been preaching in the converted. You 62 00:04:23.360 --> 00:04:28.399 mentioned a few statistics for example, about safety, about price during the meeting. 63 00:04:28.439 --> 00:04:31.519 It may not be fully understood by the general auto bio public at large. 64 00:04:32.360 --> 00:04:36.519 Yeah, so a lot of people still think teslas are super expensive because 65 00:04:36.519 --> 00:04:42.600 we did start out with an expensive sports car, then slightly less expensive sedan 66 00:04:43.040 --> 00:04:46.319 and suv. But now we're at the point where the starting price of a 67 00:04:46.399 --> 00:04:49.439 Tesla is actually below the average selling price of a car in the United States. 68 00:04:49.839 --> 00:04:56.399 So it Teslas are actually much more affordable than people realize, and so 69 00:04:57.439 --> 00:05:00.079 we should just make sure people at least know that right and that it tells 70 00:05:00.240 --> 00:05:04.480 is also the safest cars in the road US in so many ways that people 71 00:05:05.439 --> 00:05:10.040 actually a lot of people with Tesla don't even know. Um, you know, 72 00:05:10.120 --> 00:05:13.199 sort of the cabin overheat protection for example, or the fact that we 73 00:05:13.279 --> 00:05:20.040 continuously update the automatic emergency breaking software and the way the airbags deploy. So 74 00:05:20.399 --> 00:05:23.720 we look, we look at an accident, we say, is there anything 75 00:05:23.720 --> 00:05:26.759 we could do from a software standpoint to improve the cars that are already on 76 00:05:26.800 --> 00:05:30.560 the road, And then what can we update in the design to improve safety? 77 00:05:30.279 --> 00:05:36.439 So, UM, and I think that you know, the statistics speak 78 00:05:36.439 --> 00:05:40.720 for themselves. This is not simply a matter of opinion. It's the statistically 79 00:05:40.720 --> 00:05:44.160 it's safer than anything else. Um, you mentioned affordability. It's something you 80 00:05:44.240 --> 00:05:47.120 discussed during your meeting. You've also discussed it a bit on the call on 81 00:05:47.160 --> 00:05:51.199 the earnings call recently, as well, in relation to the Federal Reserve raising 82 00:05:51.199 --> 00:05:55.920 interest rates and obviously the monthly cost for a typical auto buyer, you think 83 00:05:55.920 --> 00:05:58.480 it's going to be a tough twelve months ahead. You've said that a few 84 00:05:58.519 --> 00:06:03.480 times, yes before everyone not just has understood. So it's um, I 85 00:06:03.519 --> 00:06:08.439 mean simply, uh, you know you can you can think of raising the 86 00:06:08.439 --> 00:06:11.839 FRED rate as as somewhat of a brake pedal and the economy frankly, it's 87 00:06:11.839 --> 00:06:16.600 so it's um. It makes a lot of things more expensive, certainly things 88 00:06:16.600 --> 00:06:20.000 that are bought with credit, but then it has downstream effects on on even 89 00:06:20.000 --> 00:06:24.160 things that are on board with credit. So um. You know, if 90 00:06:24.199 --> 00:06:28.920 if the car payment or your home mortgage payment is absorbing more of your monthly 91 00:06:28.959 --> 00:06:31.560 budget, then you have less to money to buy other things. So actually 92 00:06:31.759 --> 00:06:35.040 it affects everything, even those that aren't things that are important on credit. 93 00:06:35.560 --> 00:06:41.160 So UM. And my concern with the with the way that the Federal Reserve 94 00:06:41.240 --> 00:06:45.160 is making decisions is that they they're just operating with um too much latency. 95 00:06:45.240 --> 00:06:48.600 Basically, the data is somewhat stale, so they So the federals was was 96 00:06:48.800 --> 00:06:55.199 slow to raise interest rates um and and now I think they are slow to 97 00:06:55.240 --> 00:06:58.360 they're going to be slow to lower them. That's that that appears to be 98 00:06:58.439 --> 00:07:00.519 the case. Yeah, that may well be. We spent a lot of 99 00:07:00.519 --> 00:07:03.120 time talking about that is a bad imagine, perhaps too much. But when 100 00:07:03.160 --> 00:07:08.040 it comes to latency, that takes me to pricing, because you've discussed the 101 00:07:08.079 --> 00:07:11.800 lack of latency in your own ability to understand exactly what's going on in the 102 00:07:11.800 --> 00:07:16.439 market for those cars as opposed to Yeah, many of the legacy and other 103 00:07:16.600 --> 00:07:23.000 Yes. Yes, we have real time information on demand so um, so 104 00:07:23.079 --> 00:07:26.319 we know how many people place an order for it has yesterday, so the 105 00:07:26.519 --> 00:07:31.879 computer calculates that all um. And literally every day we get a you know, 106 00:07:32.079 --> 00:07:38.199 an automated email to the exact staff that says how many people place an 107 00:07:38.279 --> 00:07:42.600 order in which countries for which cars, So we know what demand, we 108 00:07:42.639 --> 00:07:46.720 know, we know what the orders were yesterday. Um. And you don't 109 00:07:46.720 --> 00:07:50.199 want to overreact to these things because sometimes you get like little dips for you 110 00:07:50.240 --> 00:07:53.680 know, reasons that are hard to explain. Do you look at them every 111 00:07:53.759 --> 00:07:57.439 day? Yeah? Yeah, um, But like I said, you don't 112 00:07:57.439 --> 00:08:00.639 want to over react to to you know, if like the week is slow 113 00:08:00.720 --> 00:08:05.360 or something, you don't overreact to that. Um. But if but if 114 00:08:05.360 --> 00:08:09.480 you look at the trend, say for the over two weeks span or something 115 00:08:09.519 --> 00:08:13.680 like that, you can see that, Okay, there's this for some reason 116 00:08:13.560 --> 00:08:20.600 the demand is is less than it was or it's higher. Um, just 117 00:08:20.839 --> 00:08:24.120 pricing. Yeah, yeah that I mean is that Are you almost like an 118 00:08:24.120 --> 00:08:28.519 airline at this point in terms of kind of dynamic pricing model? Yes, 119 00:08:28.959 --> 00:08:37.879 so, um, we we're basically adjust pricing to match demand. Um. 120 00:08:37.399 --> 00:08:41.480 And um we obviously did a big price drop in Q one, but now 121 00:08:41.879 --> 00:08:46.200 January is is usually a terrible time for car binding. Um, so there's 122 00:08:46.519 --> 00:08:50.399 there's seasonality to car purchases. Um with January January, January is often the 123 00:08:50.399 --> 00:08:56.480 worst month. So um, so we did, we did a big price 124 00:08:56.519 --> 00:09:00.480 drop and then and then then recently we did we did a pricing care So, 125 00:09:01.519 --> 00:09:07.159 as I mentioned to the audience, the realities that all companies do significant. 126 00:09:07.240 --> 00:09:11.440 All car companies make significant adjustments to price because you've got the MSRP number. 127 00:09:13.240 --> 00:09:16.799 And then if demand is high, a dealers will charge some premium of 128 00:09:16.000 --> 00:09:22.960 MSRP. If demand is lower, they will have manufacturing incentives, so that 129 00:09:22.039 --> 00:09:26.200 you can actually see a very big difference over the course of say six months 130 00:09:26.720 --> 00:09:31.919 between the peak to trough of all cars. It's just that Tesla is so 131 00:09:31.039 --> 00:09:37.080 immediate and obviously transparent. It's not a question of MSRP and then mock ups 132 00:09:37.159 --> 00:09:41.759 or discounts. Right, But after your last earnings, a lot of investors 133 00:09:41.759 --> 00:09:45.320 came away wanting to talk about your margins and wanting to talk about your pricing 134 00:09:45.639 --> 00:09:48.159 in particular. I think you know, you talked about vehicle cost reduction. 135 00:09:48.720 --> 00:09:54.399 You sort of even said it's hard to explain the profundity of technically selling cars 136 00:09:54.399 --> 00:10:00.879 now for no profit and still yielding tremendous economics over time. Yes, I'd 137 00:10:00.919 --> 00:10:03.559 like you to explain a little bit more of what that means. You did 138 00:10:03.600 --> 00:10:05.759 a bit in the meeting as well. You're talking about sort of the move 139 00:10:05.840 --> 00:10:11.639 I think from manual to autonomous and the value add that will come along with 140 00:10:11.679 --> 00:10:16.480 that. But explain to people why that's profound and why conceivably not that you 141 00:10:16.519 --> 00:10:20.360 want to You could sell automobiles now for no profit and still make enormous profit 142 00:10:20.360 --> 00:10:26.200 in the future. Yeah, So Tesla is the only car company selling cars 143 00:10:26.240 --> 00:10:31.879 that where we believe the car is capable of achieving full autonomy with a software 144 00:10:31.919 --> 00:10:37.840 update. So the value of a folios or folio autonomous car is we think 145 00:10:39.960 --> 00:10:45.840 perhaps five times more valuable than an autonomous car. Why, well, the 146 00:10:45.919 --> 00:10:48.600 utility of a car, typically a passenger car is going to be maybe ten 147 00:10:48.720 --> 00:10:52.440 ten hours a week, maybe twelve if you say, like somebody's going to 148 00:10:52.559 --> 00:10:58.080 drive an hour and a half a day on average, so maybe an hour 149 00:10:58.159 --> 00:11:01.639 commute per day and then an occasional long trip. But figure it's like ten 150 00:11:01.720 --> 00:11:07.200 twelve hours a week is typical for um, a passenger vehicle. UM. 151 00:11:09.759 --> 00:11:13.360 And then uh, you also have a lot of costs associated with parking. 152 00:11:13.360 --> 00:11:15.600 You need a garage or we've got to buy a parking space, or you've 153 00:11:15.600 --> 00:11:18.799 got to get a you know, got a parking ticket at the at the 154 00:11:18.799 --> 00:11:20.919 mole. It's there's there's a lot of costs associated with cars. UM. 155 00:11:22.440 --> 00:11:26.120 And uh, Now if you've got a car that's autonomous, UM, that 156 00:11:26.159 --> 00:11:31.919 can go around and essentially be like an autonomous uber, UM, the utility 157 00:11:33.039 --> 00:11:37.879 I think is going to be what it's going to be much higher perhaps you 158 00:11:37.879 --> 00:11:39.919 know. And then this again there was there's so much speculative I understand we're 159 00:11:39.919 --> 00:11:43.039 talking about robot taxis here at least what people have called and you have called 160 00:11:43.159 --> 00:11:48.480 Yeah, autonomous uber is what you think about. So perhaps the utility there 161 00:11:48.480 --> 00:11:50.000 would be on the order of fifty hours a week. It's just a guess, 162 00:11:50.279 --> 00:11:54.159 say there's one hundred and sixty eight hours a week. Um, and 163 00:11:54.279 --> 00:11:58.960 probably as rough guess an autonomous car is will be able to be active instead 164 00:12:00.039 --> 00:12:03.919 of four or four ten hours a week, probably in argue for about fifty 165 00:12:03.440 --> 00:12:07.759 But it's the same car, so the and it costs the same to build. 166 00:12:09.320 --> 00:12:13.240 Um, you're I want to understand the business model a little bit because 167 00:12:13.480 --> 00:12:18.279 I'm buying the car and instead of it parking at my lot while I'm working, 168 00:12:18.399 --> 00:12:20.279 it goes off and picks people up and drops them off. Yes, 169 00:12:20.759 --> 00:12:24.320 who's making the money from that? I assume that that's the value add you're 170 00:12:24.360 --> 00:12:28.559 talking about. Is it a revenue share? Do you have this model sort 171 00:12:28.559 --> 00:12:31.559 of took planned out specifically? Is how it would look? Um? Yeah, 172 00:12:31.559 --> 00:12:35.519 it's it's it's been in the tens of terms and conditions for for quite 173 00:12:35.519 --> 00:12:39.759 a long time. It has. Yeah, So the the owner of the 174 00:12:39.759 --> 00:12:43.759 car would make I don't know some amount, who knows what it'd be, 175 00:12:43.799 --> 00:12:50.080 but perhaps uh, you know, it could be fifty fifty split or seventy 176 00:12:50.120 --> 00:12:54.000 thirty. I don't know, but um, the cars are are if you 177 00:12:54.000 --> 00:12:56.879 buy it tells the car, it can only be used in to tells a 178 00:12:56.879 --> 00:13:01.919 network. It cannot be used in someone else's network. So um that that 179 00:13:01.960 --> 00:13:05.639 means that if the car is able to be used five times as much and 180 00:13:05.799 --> 00:13:15.120 and Tesla is likely to make basically two or three times the original value or 181 00:13:15.159 --> 00:13:20.320 sale value of the car um in in Robotaxi revenue. Right, this is 182 00:13:20.440 --> 00:13:26.919 this is gigantic, um ass. It'll be like selling cars for software margins 183 00:13:26.320 --> 00:13:33.600 because in fact it is software, right. So um, instead of effectively 184 00:13:33.600 --> 00:13:39.120 having say twenty five cent margins, it might be seventy seventy percent or more. 185 00:13:39.720 --> 00:13:48.159 Um and I mean the free cashal associated with that. It is actually 186 00:13:48.159 --> 00:13:52.919 truly a staggering amount. The best analysis that I've seen thus far about this 187 00:13:52.120 --> 00:13:56.159 is from Kathy Woods from arc in Best. We know Kathy, well, 188 00:13:56.279 --> 00:14:00.679 yeah, she's great. Ye I would assume you like her obviously, she's 189 00:14:00.679 --> 00:14:03.679 been a used proponent of your stock for many years. Was right, by 190 00:14:03.679 --> 00:14:09.320 the way, very early on she was right yeah. Um, So I 191 00:14:09.360 --> 00:14:11.679 thought she was being optimistic, but I she turned out to be in fact 192 00:14:13.120 --> 00:14:18.279 almost spot on. Yeah. Um And I am familiar somewhere with her analysis. 193 00:14:18.080 --> 00:14:22.440 What a lot of investors come back to where everybody is your promises of 194 00:14:22.840 --> 00:14:26.919 full autonomy. Um, and you seem to be saying again it's close. 195 00:14:28.879 --> 00:14:31.200 Yeah, you said it before, you know that. Yes, I mean 196 00:14:31.279 --> 00:14:37.639 you said nineteen, I think about twenty and I've certainly been Um. You 197 00:14:37.679 --> 00:14:41.120 know, if I guess I have somewhat of a pathological optimism. Um, 198 00:14:41.159 --> 00:14:43.039 well, it works, by the way, because you don't end up with 199 00:14:43.120 --> 00:14:46.159 this unless you're optimistic. But yeah, exactly. I mean, who would 200 00:14:46.159 --> 00:14:50.519 try to do this and this and rockets if they weren't pathologically optim No, 201 00:14:50.960 --> 00:14:54.720 you'd have to be. Yeah, so there's a you know, there's a 202 00:14:54.759 --> 00:14:56.360 double edged sword. I guess. Um. But but you know, do 203 00:14:56.399 --> 00:14:58.759 you ever say to yourself, all right, I'm getting a little ahead of 204 00:14:58.759 --> 00:15:01.039 mysel self here. I mean when you said again, we're going to have 205 00:15:01.039 --> 00:15:03.679 full autonomy. No, I didn't say, well, I said, I 206 00:15:03.720 --> 00:15:07.399 think you think, my opinion, we probably will. Okay, Right, 207 00:15:07.440 --> 00:15:11.679 there's a number of qualifiers there. I think my opinion, we probably will. 208 00:15:11.759 --> 00:15:16.240 So what do you think in your opinion you probably will win? I 209 00:15:16.279 --> 00:15:18.679 mean, it does look like it's going to happen this year. Why, 210 00:15:20.080 --> 00:15:24.399 well, we're not at the point where the car can drive on highways and 211 00:15:24.759 --> 00:15:35.279 entities with and where a human intervention is extremely rare. So I mean, 212 00:15:35.320 --> 00:15:41.200 just I was able to drive for several days just dropping a navigation pen in 213 00:15:41.279 --> 00:15:45.000 random locations in the Greater Wilson area with no interventions, and the same in 214 00:15:45.080 --> 00:15:46.799 San Francisco, which is a very difficult place to drive. So I mean 215 00:15:46.799 --> 00:15:54.080 it's you've got bus lanes, one way streets, you know, so it's 216 00:15:54.159 --> 00:15:58.480 quite quite a challenging a homeless situation. You were driving recently in San Francisco 217 00:15:58.559 --> 00:16:00.600 where the car was. I have been doing doing that quite a lot because 218 00:16:00.600 --> 00:16:07.200 of the Twitter's headquartered quarter. Maybe less, maybe a little less soon. 219 00:16:07.799 --> 00:16:11.440 Yeah, well we'll get to that. But go finish your point though, 220 00:16:11.480 --> 00:16:17.639 in terms of why you feel more, why you're optimistic for that reason, 221 00:16:17.639 --> 00:16:22.360 which is that the I've actually never, um, the first time I've had 222 00:16:22.360 --> 00:16:30.840 a situation where there's been several days in a row going picking random destinations and 223 00:16:30.960 --> 00:16:36.840 having no safety related intervention, right, so the first time ever. Um. 224 00:16:37.200 --> 00:16:38.720 I want to get to a lot of other topics, but you know, 225 00:16:38.759 --> 00:16:44.440 I am curious on Tesla in particular. UM. One thing I haven't 226 00:16:44.480 --> 00:16:48.919 heard you discuss is the impact on your business of the Inflation Reduction Act. 227 00:16:48.440 --> 00:16:52.039 At least I haven't heard it. Maybe you've discussed it. Um. You 228 00:16:52.080 --> 00:16:55.200 know Adam Jonas who follows your company. Morgan Stanley said, we believe the 229 00:16:55.200 --> 00:17:00.600 IRA is so advantageous to Tesla in terms of absolute dollars versus a legacy AMS 230 00:17:00.639 --> 00:17:03.599 it could equal as much as forty five percent of current earnings per share. 231 00:17:04.200 --> 00:17:07.799 Is the IRA going to be usually beneficial to your company? I think I'm 232 00:17:07.799 --> 00:17:14.599 not sure it'd be it's gonna be that beneficial. Um Uh, there's UM. 233 00:17:15.920 --> 00:17:21.039 There are. There are significant advantages for having domestic production of batteries, 234 00:17:21.720 --> 00:17:29.359 which I think is important for strategic reasons. You know. I think as 235 00:17:29.359 --> 00:17:33.440 we moved to an electric transport economy, I think it's important for the United 236 00:17:33.480 --> 00:17:38.119 States to UM have some independence with respect to battery production, including all of 237 00:17:38.119 --> 00:17:42.359 the precursors necessary for a battery. So the the IRA strongly instands that, 238 00:17:44.119 --> 00:17:52.279 UM, and I think that will prove to be something of important of national 239 00:17:52.279 --> 00:17:56.920 strategic importance in the future. So I'm not sure I would say it's as 240 00:17:56.119 --> 00:18:00.960 good as say a forty five percent increase in earnings to share or it's it's 241 00:18:02.000 --> 00:18:04.799 a it's it's helpful. Yeah, yeah, I would say it's maybe it's 242 00:18:04.839 --> 00:18:10.400 a half that or something, Okay, but I mean it is quite significant. 243 00:18:10.400 --> 00:18:11.640 Have you mapped to that at all? Do you have a sense. 244 00:18:11.640 --> 00:18:15.279 I mean, obviously you talk about storage, for example, which is so 245 00:18:15.359 --> 00:18:18.880 important. You spend time on that as well. The megapac Is it beneficial 246 00:18:18.920 --> 00:18:22.799 in that particular area beyond obviously the credits, the tax credits that certain people 247 00:18:22.839 --> 00:18:26.000 get for buying a Teslam. Yeah, there's two. There's really two posts 248 00:18:26.160 --> 00:18:34.079 to the IRA. There's the incentive for battery manufacturing um. And it's really 249 00:18:34.160 --> 00:18:37.799 quite a detail. It's it's actually a very well written uh well in that 250 00:18:37.920 --> 00:18:41.400 it's it really makes sure that you can't game it. You know, so 251 00:18:41.440 --> 00:18:45.440 you actually have to build the batteries in the US, and you actually have 252 00:18:45.440 --> 00:18:48.640 to build the precursis to the batteries. Um. But if you do it 253 00:18:48.799 --> 00:18:55.240 is um uh. I believe thirty dollars at the at the cell level and 254 00:18:55.440 --> 00:18:59.519 fifteen dollars at the pack level. I believe, if article correctly so, 255 00:19:00.440 --> 00:19:04.759 that that that is very significant for batteries and then you've got the consumer tax 256 00:19:04.839 --> 00:19:11.000 credit for evs provided they are built in the US. Uh and end that 257 00:19:11.079 --> 00:19:15.319 the package built in the US. So UM, no, this this this 258 00:19:15.400 --> 00:19:19.640 is the really the first time that we've we've actually had the the first time 259 00:19:19.680 --> 00:19:23.799 in many years that Tesla has actually had the consumer tax credit. Yes, 260 00:19:25.079 --> 00:19:29.480 because we actually exceeded the limit for the consumer at tax credit several years ago. 261 00:19:29.960 --> 00:19:33.119 Um. And so up until December until January this year, I should 262 00:19:33.160 --> 00:19:37.440 say, there was no consumer attax in center to buy a Tesla, but 263 00:19:37.680 --> 00:19:42.079 there was for the other car companies because they have not reached They've not made 264 00:19:42.079 --> 00:19:45.440 that many electric cars, right, they didn't reach their limit. Yeah. 265 00:19:45.559 --> 00:19:48.440 Um. You know, before I wrap on Tesla, I would like to 266 00:19:48.480 --> 00:19:51.920 get to China, which is obviously an important market for you. Um. 267 00:19:51.960 --> 00:19:55.039 You know, I noticed that chart you put up by D's the only company 268 00:19:55.079 --> 00:19:57.920 that seems to be making money so far. Um, are you concerned at 269 00:19:57.920 --> 00:20:02.680 all about U, you know, ruthless pricing sort of at the low end 270 00:20:02.680 --> 00:20:04.519 in the Chinese market and what that will mean to your market share? Though? 271 00:20:06.640 --> 00:20:11.200 Um? Well, currently a limitation UM with is that is the is 272 00:20:11.240 --> 00:20:18.359 the production rate of our Shanghai factory it's not demand so UM. I mean, 273 00:20:19.799 --> 00:20:29.359 so that's there's some constraints on our ability to expand in China and UM, 274 00:20:29.799 --> 00:20:32.200 and so it's we're making as many card as we can. It's not 275 00:20:32.240 --> 00:20:36.400 a demand issue, and you can be making batteries soon as well. Yes, 276 00:20:37.319 --> 00:20:41.799 yes, UM for the Mega pac UM for for mostly for non US 277 00:20:41.880 --> 00:20:47.079 markets. Yeah. So are you concerned at all about the growing belligerence between 278 00:20:47.119 --> 00:20:49.519 China and the US? UM. I think that should be a concern for 279 00:20:49.559 --> 00:20:55.480 everyone. I think you're right. I think it is shared by many people 280 00:20:55.519 --> 00:20:59.720 who run large organizations and smaller ones. Do you think, for example, 281 00:21:00.279 --> 00:21:07.079 wild will make a move to take control of Taiwan The official officially, the 282 00:21:07.079 --> 00:21:15.119 official policy of China is that Taiwan should be integrated. One does not need 283 00:21:15.200 --> 00:21:19.519 to read between the lines. One consintuely read the lines. Do you think 284 00:21:21.039 --> 00:21:25.960 so. I think there's a certain there's some inevitability to the situation that would 285 00:21:26.000 --> 00:21:32.000 not be good for Tesla conceivably, or for any any company in the world. 286 00:21:32.119 --> 00:21:36.000 Frankly, yes, or any company in the world. I think most 287 00:21:36.480 --> 00:21:42.039 almost no one really realizes that the Chinese economy and the global the rest of 288 00:21:42.039 --> 00:21:47.839 the global economy are like conjoin twins, it would be like trying to separate 289 00:21:47.920 --> 00:21:52.279 conjoin twins. That that's the severity of the situation. UM. And it's 290 00:21:52.279 --> 00:21:57.920 actually worse for for a lot of other companies than it is for for Tesla. 291 00:21:59.000 --> 00:22:00.160 I mean, I'm not sure sure where you're gonna get an iPhone, 292 00:22:00.200 --> 00:22:04.680 for example. UM, and Apple's recently started doing some some sort of small 293 00:22:04.680 --> 00:22:07.960 amount of production in India. But it's tiny con tiny not to mention an 294 00:22:07.960 --> 00:22:12.920 advanced semiconductor chip if they take over Taiwan Semi. Correct, you design your 295 00:22:12.920 --> 00:22:17.400 own chips, but you manufacture them at Taiwan Semi too, Right, we 296 00:22:17.480 --> 00:22:22.000 do some we do. We do. We use Samsung an TSMC UM. 297 00:22:22.640 --> 00:22:29.640 But uh, you seem to think it's it's likely to happen. I'm only 298 00:22:29.680 --> 00:22:33.680 saying that that is their policy, and I think you should take the word 299 00:22:33.759 --> 00:22:40.480 seriously. They mean it. Um. You know your time. I think 300 00:22:40.480 --> 00:22:42.279 you've said, in fact some of the brief conversations we've had, you know, 301 00:22:42.319 --> 00:22:45.599 it's one of the most valuable things in your own control. And I 302 00:22:45.640 --> 00:22:49.839 am curious now and Tesla investors sort of seemed to rejoice at the announcement of 303 00:22:49.839 --> 00:22:52.480 a new CEO at Twitter in part because they thought, well, he's going 304 00:22:52.519 --> 00:22:56.920 to have more time for Tesla. That's true, Is it true? How 305 00:22:56.960 --> 00:23:00.920 do you see sort of allocating your time now that you will suit no longer 306 00:23:00.960 --> 00:23:07.000 be the CEO of Twitter? Um? Yeah, so, um, you 307 00:23:07.039 --> 00:23:10.440 know it was. It was really very pretty challenging, um situation at the 308 00:23:10.440 --> 00:23:14.920 point which the Twitter A position closed because um, you know, in rough 309 00:23:15.039 --> 00:23:18.599 terms, uh, that there was a billion and a half dollars of debts 310 00:23:18.640 --> 00:23:22.960 osing was added. Um. While at the same time, um, there 311 00:23:22.000 --> 00:23:26.519 was a massive psyche exactly you've said you were four months away from bankruptcy. 312 00:23:26.759 --> 00:23:30.599 It was three billion dollars in a year, conceivably negative three billion dollars cash 313 00:23:30.599 --> 00:23:32.680 flow, and had a billion billion dollars in the bank. Yeah. Um, 314 00:23:32.880 --> 00:23:34.880 no, that's not good. No, anybody who knows anything about corporate 315 00:23:34.920 --> 00:23:38.640 finance knows it's bad. No. I mean I say that. The analogy 316 00:23:38.680 --> 00:23:41.400 I was using was it's like being teleported into a plane that's in a nose 317 00:23:41.480 --> 00:23:45.920 dive, head to the ground, with the engines on fire and the controls 318 00:23:45.920 --> 00:23:49.759 don't work. Right, you were that's what I want to be. Chose 319 00:23:49.759 --> 00:23:53.680 to get into the plane? Well, well I did try to exit the 320 00:23:53.720 --> 00:23:56.519 deal, and it wouldn't let me. I mean it was it was a 321 00:23:56.559 --> 00:24:00.680 funny situation where where I first, first we posed the acquisition, they said 322 00:24:00.720 --> 00:24:07.279 hell though they adopted a poison pill, basically saying they'd rather die than be 323 00:24:07.799 --> 00:24:14.240 then be acquired, like the rather chew on Sanide. But then a few 324 00:24:14.240 --> 00:24:15.920 months later that they're like, basically have a gun to my head saying no, 325 00:24:17.000 --> 00:24:18.599 you must acquire us. I mean that's quite a change. It is 326 00:24:18.680 --> 00:24:22.599 quite a change. But you seem to think I can figure out a way 327 00:24:22.599 --> 00:24:26.000 out of this, until finally, I guess you realized I can't. It 328 00:24:26.000 --> 00:24:29.480 was made clear to me that the decision would not come down in my favorite 329 00:24:29.519 --> 00:24:34.359 Leaguely yeah, all right, well now you have a new CEO. Why 330 00:24:34.519 --> 00:24:38.640 I think little Linda Yakuen is gonna be great? Why? Um well, 331 00:24:41.279 --> 00:24:45.519 Twitter is it's very much an advertising dependent business. Linda's obviously incredible at bat 332 00:24:45.559 --> 00:24:51.599 and she's just a great executive in general. Um So, the and you 333 00:24:51.640 --> 00:24:57.640 know, my skills and interests are in technology. Um so, you know 334 00:24:57.640 --> 00:25:06.480 I'll continue to play role advancing the software. Um and you know, you 335 00:25:06.559 --> 00:25:11.160 know, getting the futures product stuff basically. Um So, I mean the 336 00:25:11.279 --> 00:25:17.440 general idea is Linda will operate the company and I will build products. Yeah, 337 00:25:17.599 --> 00:25:18.759 and that's I think that's that's that's a good, that's a good, 338 00:25:18.799 --> 00:25:22.599 and that will take less of your time. Yeah, and I have I 339 00:25:22.680 --> 00:25:27.559 have that situation at SpaceX. Um you know, Gwynchotwell, who's amazing um 340 00:25:29.039 --> 00:25:33.440 uh, you know, operates the company and I and I booked the sort 341 00:25:33.440 --> 00:25:36.640 of advanced technology, right, both sort of designing in new rockets and that 342 00:25:36.720 --> 00:25:40.000 kind of thing. And and then that's uh, you know, everyone you 343 00:25:40.119 --> 00:25:41.680 played to their strengths and that's that's that's mine. That's although I wonder you 344 00:25:41.720 --> 00:25:45.359 mean in twenty twenty one you said you spend most of your time on the 345 00:25:45.400 --> 00:25:48.480 development of Starship. I think you've also been quoted I've read I think I 346 00:25:48.559 --> 00:25:52.920 heard you say, actually think most of your mental energy at certain points has 347 00:25:52.960 --> 00:25:56.680 been on Starship. Yes, So I wonder have you been diverted from Starship 348 00:25:56.680 --> 00:26:00.640 a little bit and we'll returned to it as the main focus or is it 349 00:26:00.720 --> 00:26:03.880 Tesla? You know again, your time, I know, having tried to 350 00:26:03.920 --> 00:26:07.720 schedule this interview with time. Time treach is a real thing. From Yeah. 351 00:26:08.200 --> 00:26:14.119 In fact, I I'd say that the only real currency is time. 352 00:26:14.720 --> 00:26:17.240 Time is the only true currency. So where you're going to spend the currency 353 00:26:17.279 --> 00:26:21.160 now that you don't have to spend as much a Twitter, Well, I'm 354 00:26:21.160 --> 00:26:27.279 going to be devoting a lot more time to to Tesla UM and especially on 355 00:26:27.519 --> 00:26:36.599 the AI development UM and your product development, the and and then there's and 356 00:26:36.799 --> 00:26:42.000 I'll also be allocating some more time to getting Starship to orbit UM. In 357 00:26:42.000 --> 00:26:47.599 the case of SpaceX UM, what I was able to do there was to 358 00:26:48.720 --> 00:26:53.680 as as the Starlink constellation I started working. I was able to move some 359 00:26:53.759 --> 00:27:00.039 of the best people from the Starlink program to the Starship program UM, so 360 00:27:00.599 --> 00:27:04.279 they're you know, the timing kind of worked out well that that UM. 361 00:27:06.480 --> 00:27:10.039 I mean, we have really some of the world's best engineers at SpaceX and 362 00:27:10.079 --> 00:27:14.319 Tesla, really some incredible human beings. There was some statistic you shared. 363 00:27:14.559 --> 00:27:18.039 I think three point six million people applied for a job at Tesla. I 364 00:27:18.039 --> 00:27:19.200 think I didn't I wrote it down. I didn't write it down here since 365 00:27:19.200 --> 00:27:22.720 a member, but that was that the number. Yeah, it's crazy, 366 00:27:23.519 --> 00:27:27.519 there's a lot of demand. Have one hundred and twenty seven almost one undred 367 00:27:27.519 --> 00:27:32.440 twenty thousand employees, but three point six million people applying for a job. 368 00:27:33.759 --> 00:27:37.920 Yes, and I mean there's you know, we would only add like to 369 00:27:37.680 --> 00:27:41.759 twenty or thirty thousand jobs. So it's it's it's sort of the acceptance rate 370 00:27:41.799 --> 00:27:48.680 for for Tesla is it's much more difficult to get into Tesla or SpaceX than 371 00:27:48.759 --> 00:27:52.880 Harvard, right, the acceptance rate is it's even lower one for two. 372 00:27:53.400 --> 00:27:56.480 Well, yeah, it's it's it's it's the acceptance rate is lower than the 373 00:27:56.480 --> 00:28:03.759 most demanding of uses in the world. It's insane. Speaking of employment, 374 00:28:03.799 --> 00:28:07.079 though, you had seventeen hundred people at Twitter when the plane was nose diving, 375 00:28:07.160 --> 00:28:11.359 I think you're a fifteen hundred. Now, that's roughly sixty three hundred 376 00:28:11.440 --> 00:28:17.400 people. Were they all superfluous? No, not all. Do you figure 377 00:28:17.440 --> 00:28:18.720 out which ones are or is it a little late? You know, sometimes 378 00:28:18.720 --> 00:28:22.920 it gets a little late. I mean, looked desperate time was called for 379 00:28:22.200 --> 00:28:27.799 desperate measures. So um, there's no question that some of the people who 380 00:28:27.799 --> 00:28:33.200 will let go probably shouldn't have been let go because we we simply did not 381 00:28:33.319 --> 00:28:37.559 have the time to figure out we had to be you know, make make 382 00:28:37.720 --> 00:28:42.720 widespread cuts to get the run rate under control. So um, you know, 383 00:28:42.759 --> 00:28:47.039 and uh so this is not to say that hey, everyone who is 384 00:28:47.039 --> 00:28:51.440 like going from Twitter is like somehow terrible or something. It's just we have 385 00:28:51.519 --> 00:28:57.359 to with very little information. Uh, get get the the the headcount expenses, 386 00:28:57.359 --> 00:29:02.880 and that the non Postonalexy it is down to or at least break even, 387 00:29:03.359 --> 00:29:07.119 And we're not quite a break even yet, but we're close. We 388 00:29:07.160 --> 00:29:08.880 need to do it fast and if you, if you, if you do 389 00:29:08.920 --> 00:29:12.640 it fast. Unfortunately, there are gonna be some babies thrown out with the 390 00:29:12.680 --> 00:29:17.799 South orders. So um so, I definitely, I definitely would not want 391 00:29:17.839 --> 00:29:22.160 to say, uh, you know, disparage. You know, any I 392 00:29:22.200 --> 00:29:23.759 understands. But it's funny, you know, I hear from other I've heard 393 00:29:23.759 --> 00:29:29.640 from other tech CEOs quietly they look at what you did at Twitter, and 394 00:29:29.640 --> 00:29:33.200 they sort of they won't admit it publicly, but they, you know, 395 00:29:33.240 --> 00:29:36.599 they said, wow, that was something. I mean to cut that deeply 396 00:29:36.680 --> 00:29:38.880 and still be able to shun the company. People will argue about how well 397 00:29:40.000 --> 00:29:42.599 or things were running. Have you heard from any of the CEOs said thank 398 00:29:42.599 --> 00:29:47.240 you for doing that giving us sort of runway or leeway to at least make 399 00:29:47.279 --> 00:29:49.680 some cuts of our own. Um I have I've heard that from a few 400 00:29:49.720 --> 00:29:53.839 people and and I've heard that through the grapevine. Yeah. Um, but 401 00:29:53.960 --> 00:29:59.880 um, like I said, if desperate times call for desperate measures, or 402 00:30:00.200 --> 00:30:02.920 that those were desperate times. So do you see starting to I mean, 403 00:30:02.920 --> 00:30:04.599 we'll lend hire people. Do you think are you ready for people to be 404 00:30:04.640 --> 00:30:08.559 hired to Twitter? You know, I think we absolutely um need to need 405 00:30:08.559 --> 00:30:12.160 to hire people and and and if they're not too mad at us, probably 406 00:30:12.240 --> 00:30:18.079 re hire some of the people that they were let go. Um, and 407 00:30:18.119 --> 00:30:22.000 they're gonna have to come into the office. Look, I'm a big believer 408 00:30:22.480 --> 00:30:26.920 that that that that people need to are more productive when they're in person. 409 00:30:26.160 --> 00:30:33.799 Um, and and and and and really man and the whole sort of work 410 00:30:33.839 --> 00:30:41.000 from home thing, it's like I think it's Look, there are some exceptions, 411 00:30:41.039 --> 00:30:44.559 but I kind of think that that the whole notion to work from home 412 00:30:44.640 --> 00:30:48.720 is a bit like the you know, the fake Marie Antoinette quote. Let 413 00:30:48.759 --> 00:30:52.759 the meat cake. It's like it's like it's like, really, you're gonna 414 00:30:52.759 --> 00:30:55.680 work from home, and you're gonna make everyone else who made your car come 415 00:30:55.799 --> 00:30:57.319 work to the fact work in the factory. You're gonna make the people who 416 00:30:57.599 --> 00:31:00.359 make your food that gets to live it that they can't work from home. 417 00:31:00.720 --> 00:31:04.519 That you know, the people that come fix your house, they can't work 418 00:31:04.559 --> 00:31:10.599 from home, but you can. Does that seem morally right? That's messed 419 00:31:10.640 --> 00:31:12.440 up. You see there's a moral issue. Yes, I mean I see 420 00:31:12.440 --> 00:31:15.640 it more. It's a productivity issue, but it's also a moral issue. 421 00:31:17.480 --> 00:31:19.559 People should get off the moral high horse. Would work from home bullshit? 422 00:31:21.000 --> 00:31:23.519 Um because they're asking everyone else to not work from home while they do. 423 00:31:25.279 --> 00:31:27.079 And they're still pushback. By the way, it's still going on. This 424 00:31:27.160 --> 00:31:32.039 battle is still happening. I mean leaders of organizations and I speak to plenty 425 00:31:32.039 --> 00:31:34.480 of them. I want people back. I want people back three days a 426 00:31:34.480 --> 00:31:38.680 week. They're still battling. It's not clear that it's ever going to change 427 00:31:38.759 --> 00:31:45.720 to people are not coming back. People the laptop classes living in lala land. 428 00:31:45.880 --> 00:31:49.480 Okay, As I said, you can't look at the cars while people 429 00:31:49.480 --> 00:31:55.279 are working from from home here of course not um so so the people were 430 00:31:55.680 --> 00:31:59.559 you know, building cars, servicing the cars, building houses. It's fixing 431 00:31:59.640 --> 00:32:04.039 houses, making the food, um, making all the things that that that 432 00:32:04.240 --> 00:32:09.079 people consume. It's it's messed up to assume that that this they have to 433 00:32:09.079 --> 00:32:14.319 go to work, but you don't how is that that is. It's not 434 00:32:14.400 --> 00:32:20.359 just a productivity thing. I think it's Marley Marley wrong. Although productivity is 435 00:32:20.359 --> 00:32:23.640 definitely impacted too, and the ability people learn, I mean on and on 436 00:32:24.240 --> 00:32:27.880 listen, I like, so, I mean, look, you know people 437 00:32:27.880 --> 00:32:30.640 will disagree with me about this, but I you know, it's like, 438 00:32:30.519 --> 00:32:34.200 um, so if you want to work at Tesla, you want to work 439 00:32:34.200 --> 00:32:37.119 at SpaceX, you want to work at Twitter, You've got to come into 440 00:32:37.160 --> 00:32:40.359 the office every day. Yes, I mean you know, Like I'm not 441 00:32:40.400 --> 00:32:45.319 saying um, I'm saying, like, look, we put forty hours in, 442 00:32:45.359 --> 00:32:47.119 you know, and it, frankly, it doesn't even even need to 443 00:32:47.160 --> 00:32:52.640 be like, you know, Monday through Friday. You know you work Monday 444 00:32:52.680 --> 00:32:54.799 through Thursday. And also not saying no one should take I mean, I 445 00:32:54.839 --> 00:32:59.839 think people should take vacations, like I work seven days a week, but 446 00:33:00.000 --> 00:33:02.400 but I'm not expecting others to do that. How muchly do you get by 447 00:33:02.400 --> 00:33:07.599 the way, about six hours you do six hours typically? Yes, No, 448 00:33:07.759 --> 00:33:09.400 that's not bad. I thought it would be less. I've tried less, 449 00:33:09.440 --> 00:33:13.880 but my productivity, even though I'm awake more hours, I get less 450 00:33:13.880 --> 00:33:16.440 done, okay, And the brain pain level is bad if I get less 451 00:33:16.440 --> 00:33:21.559 than six hours, but you work seven days a week. Yes, yeah, 452 00:33:21.240 --> 00:33:27.920 I actually only um in terms of actually you say, like, how 453 00:33:28.000 --> 00:33:30.759 many days in a year do I not put in some meaningful amount of work? 454 00:33:30.799 --> 00:33:37.079 It's probably about two or three, right, two or three days a 455 00:33:37.200 --> 00:33:40.559 year? Yeah? A year? Yeah? Yes, Um. I want 456 00:33:40.559 --> 00:33:44.720 to get to some more specific questions about Twitter that's sort of have a more 457 00:33:45.400 --> 00:33:46.759 global aspect. Then I still want to get to AI. You're gonna give 458 00:33:46.799 --> 00:33:49.680 us some time here, can we still? Um? Yeah? I think 459 00:33:49.680 --> 00:33:52.400 We've got a board meeting coming up, but I'm happy to give as much 460 00:33:52.400 --> 00:33:54.000 time as reasonably possible. I think it's maybe it's six or something. Okay, 461 00:33:54.000 --> 00:33:59.720 Well, let's time, is it now? I love another twenty all 462 00:33:59.799 --> 00:34:02.440 right, at least twenty great. Um, let's talk about free speech a 463 00:34:02.480 --> 00:34:06.200 bit. You know, you call yourself a free speech absolutist. You want 464 00:34:06.240 --> 00:34:08.800 Twitter, and this isn't aspirationally. Aspirationally you want Twitter to be as truthful 465 00:34:08.840 --> 00:34:14.880 as possible, most accurate source of information about the world. Um, so 466 00:34:15.079 --> 00:34:19.519 what does that mean for how you police lies on the platform? You mentioned 467 00:34:19.519 --> 00:34:23.360 community notes? Is that I think community? Yeah? I mean i'd say 468 00:34:23.519 --> 00:34:32.199 so. My overall kind of vision for extra Twitter is to be a cybernetic 469 00:34:32.519 --> 00:34:37.239 collective mind for humanity. This is going to sound quite esoteric and sci fi, 470 00:34:37.400 --> 00:34:43.800 but um, so the if if you know, in pursuit of that 471 00:34:43.840 --> 00:34:49.679 objective, you want to have information move quickly, have that information be accurate, 472 00:34:49.840 --> 00:34:52.119 and you want to have error correction on that information. So you can 473 00:34:52.119 --> 00:34:55.440 think of community notes as like an error correction on information in the network. 474 00:34:57.519 --> 00:35:00.840 And the effect of community notes is actually bigger at the and it would seem 475 00:35:00.880 --> 00:35:04.559 it's bigger than the number of notes, because if somebody knows that they're going 476 00:35:04.559 --> 00:35:08.440 to get noted, they are less likely to say something that is false because 477 00:35:08.519 --> 00:35:13.880 it's embarrassing if you get community noted. Okay, and and that applies even 478 00:35:13.920 --> 00:35:20.199 to advertisers. By but we've lost that's so far forty million dollars in advertising 479 00:35:20.920 --> 00:35:28.000 because it was misleading or because because the community notes to two pretty big advertisers 480 00:35:28.000 --> 00:35:36.519 got community noted. Um. And I yes, um, I think on 481 00:35:36.639 --> 00:35:40.039 balance the community notes were correct. And I did I did say to those 482 00:35:40.039 --> 00:35:50.599 advertisers, look provide just go on Twitter and provide some facts that contradict the 483 00:35:50.639 --> 00:35:52.760 community note. That's the way to deal with the community note is to say, 484 00:35:53.519 --> 00:35:57.920 is that the community note is saying that the ad is misleading for the 485 00:35:57.920 --> 00:36:04.039 following reasons. If you've got information that rebuts that note, then just add 486 00:36:04.079 --> 00:36:07.440 that to the ad. Okay, well, we're coming up on an election. 487 00:36:07.519 --> 00:36:09.079 I mean it's a ways away, but it's going to all start. 488 00:36:09.559 --> 00:36:14.840 Um. Yeah, President Trump is allowed back on the platform. He hasn't 489 00:36:14.840 --> 00:36:16.719 actually come back, right, but one would imagine if and when he does, 490 00:36:16.840 --> 00:36:21.880 or there are others who will say twenty twenty election was rigged? Is 491 00:36:21.920 --> 00:36:27.360 that something? I assume that's not something you believe? I well, I 492 00:36:27.400 --> 00:36:34.199 think the answer The answer is is nuanced, like do I believe Biden? 493 00:36:34.320 --> 00:36:37.920 One? Yes, I believe you want you voted for I did? Actually 494 00:36:37.119 --> 00:36:47.119 do you regret that? I mean, man, I wish we could have 495 00:36:47.159 --> 00:36:51.760 just a normal human being as president. That's what I want't I think, 496 00:36:51.800 --> 00:36:53.840 if if, if we could, you know, there's that that old saying 497 00:36:53.840 --> 00:37:00.400 of like we're better we're better off being run by um, by people picked 498 00:37:00.400 --> 00:37:02.119 at random from the phone book, then the faculty of Harvard. Yeah, 499 00:37:02.199 --> 00:37:06.519 I don't know I said that, but somebody very wise, um, and 500 00:37:07.000 --> 00:37:09.199 I would I would say, if we could do that, for the president 501 00:37:09.920 --> 00:37:15.760 that that would be you're not obviously you're not happy with b don't we all 502 00:37:15.800 --> 00:37:21.039 just want an all human beings and whatever resids whatever? Sure, normal means 503 00:37:21.079 --> 00:37:23.320 not what I mean like you know, just I don't know. So you 504 00:37:23.360 --> 00:37:29.840 want somebody's competent it's helpful, Yes, I think definitely. Um, somebody's 505 00:37:29.880 --> 00:37:32.679 executive ability is underrated. Since the president is effectively the chief executive officer of 506 00:37:32.719 --> 00:37:37.880 the country. Um, it actually matters if they are a good executive officer. 507 00:37:37.960 --> 00:37:40.119 Yeah, it's it's not simply a matter of do they share your beliefs 508 00:37:40.119 --> 00:37:45.039 and you know, um, but but are they good at getting things done? 509 00:37:45.079 --> 00:37:49.159 There's a lot of decisions that need to be made every day. Um, 510 00:37:49.400 --> 00:37:52.679 many of them are unrelated to moral beliefs, you know, right, 511 00:37:53.119 --> 00:38:00.840 and and you just want a good executive because they're seeing you over America they 512 00:38:00.880 --> 00:38:04.239 are they are. We want a good CEO of America. I was going 513 00:38:04.280 --> 00:38:09.320 to do be it effective. Unfortunately, we live in highly partisan times where 514 00:38:09.480 --> 00:38:14.599 there is were about everything, including ideas, including the truth, which gets 515 00:38:14.639 --> 00:38:17.960 back to it's not true that the election in twenty and twenty was rigged. 516 00:38:19.000 --> 00:38:22.400 It wasn't stolen, and I wonder on the platform when you see that, 517 00:38:22.679 --> 00:38:24.559 does that end up in a community note or is that something you take more 518 00:38:24.599 --> 00:38:28.000 action on? And obviously there places so many I mean, to be clear, 519 00:38:28.039 --> 00:38:34.480 I don't think it was there was a stolen election. Um. But 520 00:38:34.639 --> 00:38:37.880 by the same token, if somebody's going to say that there's there's never any 521 00:38:37.920 --> 00:38:42.840 election fraud anywhere, this is obviously also false. Yeah. If if if 522 00:38:42.880 --> 00:38:46.039 one hundred million people vote, the probability that the fraud is zero is zero. 523 00:38:47.480 --> 00:38:51.920 Of course there's always going to be some right, I mean, is 524 00:38:51.960 --> 00:38:54.440 perhaps, I mean there were This election was audited, it was so many 525 00:38:54.519 --> 00:38:58.159 judges, and it went on and on and on, and there was no 526 00:38:58.480 --> 00:39:01.440 nothing whatsoever that I don't want to debate this with you. My question is 527 00:39:01.480 --> 00:39:05.760 more about what I think it's important to say, like that in any given 528 00:39:05.760 --> 00:39:07.519 election, even if you try your hardest, if you've got one hundred million 529 00:39:07.559 --> 00:39:10.159 votes, there's gonna be some some amount of fraud that is not zero, 530 00:39:10.400 --> 00:39:15.159 and that that it's important to acknowledge that without saying that that the fraud was 531 00:39:15.199 --> 00:39:21.960 of sufficient magnitude to change the outcome. So uh so, my opinion would 532 00:39:21.960 --> 00:39:23.480 be that there was some there was some small amount of fraud, but it 533 00:39:23.519 --> 00:39:25.440 was not enough to change the right and by the way, it might have 534 00:39:25.440 --> 00:39:28.480 been either way. I mean, you know, yeah, there will probably 535 00:39:28.519 --> 00:39:30.239 a little bit, but again it's gonna be You're gonna let people say that 536 00:39:30.280 --> 00:39:32.960 though on the on Twitter, and then you're gonna hope that they're corrected, 537 00:39:34.000 --> 00:39:38.239 and they will be corrected. They will be Oh yeah. Let's talk a 538 00:39:38.239 --> 00:39:44.719 bit about your tweets, um, because it comes up a lot um even 539 00:39:44.760 --> 00:39:46.840 today it came up and you know, anticipation of this. I mean, 540 00:39:47.480 --> 00:39:51.960 um, you know, you do some tweets that seem to be or at 541 00:39:52.039 --> 00:39:59.119 least give support to some who would call others conspiracy theories. Well yes, 542 00:39:59.199 --> 00:40:06.039 but I mean honestly, you know some of these conspiracy theories, uh I've 543 00:40:06.079 --> 00:40:09.920 turned out to be true. Which ones, well, like the hunter Biden 544 00:40:10.000 --> 00:40:15.800 laptop that's true. Yeah yeah, so uh you know that that that that 545 00:40:15.880 --> 00:40:20.280 was a pretty big deal. There was Twitter and and and others engaged in 546 00:40:20.320 --> 00:40:23.159 active suppression of information that was relevant to the public. M that's that's a 547 00:40:23.239 --> 00:40:27.400 that's a terrible thing that happened. That's like your interference. But how do 548 00:40:27.440 --> 00:40:30.159 you make a choice? You don't see I mean in terms of when you're 549 00:40:30.159 --> 00:40:34.639 going to engage. I mean, for example, even today, Eli you 550 00:40:34.400 --> 00:40:38.199 you tweeted this thing about George Soros. Well, I'm looking for because I 551 00:40:38.239 --> 00:40:40.400 want to make sure I quote it properly. But I mean, you know 552 00:40:40.440 --> 00:40:44.519 what you wrote, but it basically reminds me of my NATO. This is 553 00:40:44.559 --> 00:40:46.400 like, you know, calm down, people, this is not like case 554 00:40:46.400 --> 00:40:51.199 out of it. As you said, he wants to route the very fabric 555 00:40:51.239 --> 00:40:54.320 of civilization and Sos hates humanity. Like when you do something like that. 556 00:40:54.719 --> 00:40:58.239 Yeah, I think that's true. That's my opinion. Okay, but why 557 00:40:58.239 --> 00:41:02.039 share it? Why share it especially? I mean why share it when people 558 00:41:02.079 --> 00:41:06.719 who buy teslas may not agree with you, advertisers are on Twitter may not 559 00:41:06.760 --> 00:41:09.639 agree with you. Why not just say hey, I think this. You 560 00:41:09.679 --> 00:41:12.920 can tell me, we can talk about it over there, you can tell 561 00:41:12.960 --> 00:41:17.559 your friends, But why share it widely? I mean, this is a 562 00:41:17.599 --> 00:41:21.360 freedom of speech. I'm allowed to say whatever you absolutely are, But I'm 563 00:41:21.360 --> 00:41:22.719 trying to understand why you do, because you have to know it's got a 564 00:41:23.199 --> 00:41:28.360 there. It puts you in in the middle of a the partisan divide in 565 00:41:28.360 --> 00:41:30.719 the country. It makes you a lightning rod for criticism. I mean, 566 00:41:30.800 --> 00:41:35.880 do you like that. You know, people today saying he's an anti Semite. 567 00:41:35.920 --> 00:41:37.079 I don't think you are. No, I'm definitely. I'm like, 568 00:41:37.239 --> 00:41:42.559 I'm like a pro Semite. If anything, I believe that probably is the 569 00:41:42.599 --> 00:41:45.920 case. Why would you even introduce the idea of it that that would be 570 00:41:46.079 --> 00:41:49.960 the case. I mean, it looks we don't want to make this a 571 00:41:49.960 --> 00:41:52.199 George Stars interview. No, God, no, I don't. I don't 572 00:41:52.199 --> 00:41:53.639 want it at all. But I'm what I'm trying even came up there when 573 00:41:53.679 --> 00:41:58.400 the annual meeting. I mean, you know, do your tweets hurt the 574 00:41:58.400 --> 00:42:00.880 company? Are there test loan or just I don't agree with his political position 575 00:42:01.000 --> 00:42:05.280 because and I know it because he shares so much of it, or their 576 00:42:05.320 --> 00:42:08.039 advertisers on Twitter that Linda Yak Greena will come and say you gotta stop man, 577 00:42:08.280 --> 00:42:12.159 or you know, I can't get these ads because of some of the 578 00:42:12.159 --> 00:42:31.519 things you tweet. You know, I'm reminded of as we'll see it in 579 00:42:31.559 --> 00:42:38.880 the Princess Bride Great movie, Great Um, where he confronts the person who 580 00:42:38.960 --> 00:42:51.960 killed his father. He says, offer me money, offer me power. 581 00:42:52.239 --> 00:43:01.400 I don't care. See, you just don't care. You want to share 582 00:43:01.480 --> 00:43:05.119 what you have to say, I'll say I don't want to say and if 583 00:43:05.199 --> 00:43:10.760 if if the consequence of that is losing money, so be it. Okay. 584 00:43:12.519 --> 00:43:15.760 But I mean when you when you when you linked to somebody who's talking 585 00:43:15.760 --> 00:43:22.679 about the guy who killed children in a mall in Allen, Texas, you 586 00:43:22.760 --> 00:43:25.119 say something like it might be a bad siop. I'm not quite sure what 587 00:43:25.159 --> 00:43:32.960 you meant. But oh, in that particular case, Uh, there was 588 00:43:36.079 --> 00:43:44.599 a somehow that's that's not not not that the that the people people were killed, 589 00:43:44.639 --> 00:43:49.840 but the it was I think incorrectly described to be a white supremacist action. 590 00:43:50.840 --> 00:43:55.519 Um. And the evidence for that was some obscure Russian website that no 591 00:43:55.559 --> 00:44:01.719 one's everihood of that had no followers, um and the like. The company 592 00:44:01.719 --> 00:44:06.480 that that found this is belling Cat. Right. Did you know what belling 593 00:44:06.519 --> 00:44:10.599 Cat does? Psy ops? Right? I couldn't really even follow exactly what 594 00:44:10.639 --> 00:44:14.679 it was you were trying to express there, So that's part why I was 595 00:44:14.719 --> 00:44:16.679 curious. But I'm saying that I thought this, the ascribing a two white 596 00:44:16.679 --> 00:44:22.280 supremacy was bullshit, okay, and and uh and and and that the information 597 00:44:22.360 --> 00:44:28.719 for that came from an obscure Russian website and was somehow magically found by belling 598 00:44:28.760 --> 00:44:34.079 Cat, which is a company that does psy ops. And there's no proof 599 00:44:34.079 --> 00:44:37.880 by the way that he was not. There's no I would say that there's 600 00:44:37.920 --> 00:44:40.800 no proof that he is. And that's a debate you want to get into 601 00:44:40.800 --> 00:44:45.639 on Twitter. Yes, because we should not be ascribing things to white supremacy. 602 00:44:45.920 --> 00:44:51.119 If if there, if it's false. Okay, um, can we 603 00:44:51.119 --> 00:44:53.480 talk about AI now? Actually I want to talk about AI. Well, 604 00:44:53.639 --> 00:44:58.119 let me end with Twitter in the sense of Sam Altman was on the Hill 605 00:44:58.239 --> 00:45:04.880 today and he said AI's ability to manipulate interactive disinformation is a significant area of 606 00:45:04.920 --> 00:45:08.840 concern. It's moving so quick, it is a concern. And how do 607 00:45:08.880 --> 00:45:12.639 you know, I'm curious as to whether you agree with that? How you 608 00:45:12.679 --> 00:45:16.199 see that even playing out on Twitter with people who know somebody could you know 609 00:45:16.239 --> 00:45:20.679 look like you and or me and use our voice. I don't know what 610 00:45:20.760 --> 00:45:23.079 it could be. Oh that that that that that is happening in big time. 611 00:45:23.360 --> 00:45:31.920 Um so the so the reason that I'm asking people to be verified on 612 00:45:32.280 --> 00:45:37.360 Twitter and and and that we're saying, okay, verification means you've got a 613 00:45:37.360 --> 00:45:42.719 phone number from a reputable carrier, which means that you've at least passed through 614 00:45:42.800 --> 00:45:49.199 whatever they're uh security mechanisms are that you have a credit card, which so 615 00:45:49.239 --> 00:45:53.559 you now have passed through whatever security mechanisms the credit card company has. And 616 00:45:53.599 --> 00:45:59.280 that there's some small amount of money paid per month. Um uh. That 617 00:45:59.280 --> 00:46:05.559 that sort of action um uh significantly increases the cost of of fake accounts. 618 00:46:06.119 --> 00:46:13.519 Um and the with with modern with with with the latest AI um it can 619 00:46:13.559 --> 00:46:17.719 bypass basically every test for are you a human? So so then how do 620 00:46:17.760 --> 00:46:22.480 you know that a million accounts were created? How do you know that those 621 00:46:22.480 --> 00:46:28.000 were people? I don't know? How do you exactly you have to do? 622 00:46:28.079 --> 00:46:31.760 You have to do account verification um and and the thing that makes the 623 00:46:32.280 --> 00:46:36.320 like I sort of put myself in a position of like, if it was 624 00:46:36.360 --> 00:46:40.360 my goal to manipulate public opinion and create millions of accounts and make a team 625 00:46:40.400 --> 00:46:45.199 as though a topic was trending um and and that and that this is actually 626 00:46:45.199 --> 00:46:50.039 what the public beliefs um and But in order to do so, I had 627 00:46:50.039 --> 00:46:54.159 to get a million phone numbers, a million credit card numbers and pay uh, 628 00:46:54.440 --> 00:46:59.880 you know, eight bucks a month um and and and have that old 629 00:47:00.079 --> 00:47:05.880 not be traceable and clustered. I would say it's impossible. Impossible. So 630 00:47:05.920 --> 00:47:12.719 the goal of the uh of the sort of Twitter verification is fundamentally to prevent 631 00:47:12.880 --> 00:47:16.360 AI manipulation of the system. Got it. Um final question on Twitter, 632 00:47:16.920 --> 00:47:22.480 Walter Isaacson, your biographer, he said, your big goal for Twitter is 633 00:47:22.480 --> 00:47:30.320 disrupting the banking industry. Um. I'd say that's that's that's a that's a 634 00:47:30.840 --> 00:47:34.159 look. First of all, I don't want to disrupt something with the psake 635 00:47:34.199 --> 00:47:36.920 of disrupting it. It's more like, if there is a better product, 636 00:47:37.039 --> 00:47:39.519 that's great. But I'm not output disruption for the tsake disruption. I'm like, 637 00:47:39.920 --> 00:47:44.199 if we're gonna make a product that improves a quality of life for people 638 00:47:44.239 --> 00:47:50.800 that they find more useful, that that's great. Um and Uh, what 639 00:47:50.800 --> 00:47:54.639 what people see in PayPal is sort of like sort of a halfway. It's 640 00:47:54.679 --> 00:47:59.639 sort of a sort of it's frankly sort of a half baked version of what 641 00:47:59.639 --> 00:48:04.800 it could be. Um and um and so I think there's potential to create, 642 00:48:05.119 --> 00:48:12.360 um, a more efficient financial system. UM. And here we can 643 00:48:12.360 --> 00:48:15.119 get again quite quite esoteric, sort of into a sort of information theory. 644 00:48:15.119 --> 00:48:21.719 But um, the actual financial system today is a heterogeneous set of databases running 645 00:48:21.719 --> 00:48:24.800 on mainframes and coval um that still engage in batch processing. It's really quite 646 00:48:27.079 --> 00:48:31.760 uh very inefficient. Um so, UM things are still not real time um 647 00:48:31.800 --> 00:48:37.760 and and so it's possible to have um a much more efficient, homogeneous real 648 00:48:37.800 --> 00:48:45.360 time data system. Money is just information, um and and so the so. 649 00:48:45.679 --> 00:48:49.280 But that's not like the only reason. It's it's just a thing that 650 00:48:49.280 --> 00:48:53.840 that would be I think poetic to fulfill ultimately the vision that I had for 651 00:48:54.199 --> 00:49:02.119 X over twenty three years ago. Um and and actually see that counter fruition 652 00:49:02.199 --> 00:49:07.000 would be nice. But but there are many other things for Twitter, it's 653 00:49:07.039 --> 00:49:10.639 not just financial. Well, you talk about enhancing humanity. You know, 654 00:49:10.679 --> 00:49:15.440 I'm curious than a bad AI, which many people say will lead to great 655 00:49:15.480 --> 00:49:20.039 productivity gains. You show those robots, I mean, I can imagine what 656 00:49:20.079 --> 00:49:22.920 they conceivably could do empowered by AI. But I'm also curious, because you've 657 00:49:22.920 --> 00:49:28.559 certainly been concerned, what percentage do you give the chance that it will destroy 658 00:49:28.840 --> 00:49:37.360 humanity? Well, the event of artificial general intelligence is cult to singularity because 659 00:49:37.800 --> 00:49:44.119 it is so hard to predict what will happen after that. UM. I 660 00:49:44.159 --> 00:49:46.880 think it's it's it's very much a double double edged sword. I think it's 661 00:49:47.239 --> 00:49:52.480 there's there's a there's a strong probability that it will make life much better and 662 00:49:52.519 --> 00:49:57.599 that will have an age of abundance um and and there's some chance that it 663 00:49:57.639 --> 00:50:02.440 goes wrong U and destroys humanity. Hopefully that chance is small, but it's 664 00:50:02.480 --> 00:50:06.960 not zero um And so I think we want to take whatever actions we can 665 00:50:07.000 --> 00:50:09.880 think of to minimize the probability that AI goes And you've called for a pause 666 00:50:09.960 --> 00:50:13.639 along with the number of other people. Yes, I look when I call 667 00:50:13.679 --> 00:50:16.199 for the a friend of mine, Max Techmark, because a physicist said at 668 00:50:16.519 --> 00:50:21.119 mit UM you know, I wanted me to sign on to the letter. 669 00:50:21.119 --> 00:50:24.480 And it's it's like I knew it would be futile. It is futile. 670 00:50:24.639 --> 00:50:27.880 I knew would be futile. I just want to call it like it's one 671 00:50:27.920 --> 00:50:30.199 of those things. Well, for the record, I recommended that we pause. 672 00:50:30.639 --> 00:50:34.199 Uh did I think we would there would be a pause? Absolutely not. 673 00:50:36.000 --> 00:50:40.599 So you're starting, uh x AI, I think that's what you're calling 674 00:50:40.599 --> 00:50:45.760 it or some new AI effort. Um. Well, how is it going 675 00:50:45.800 --> 00:50:51.000 to be different than open AI or alphabet This is this is not the time 676 00:50:51.039 --> 00:50:57.639 to to this. I we don't have enough time and and order. Is 677 00:50:57.679 --> 00:51:01.440 this the moment to really talk about it? Um, we will have a 678 00:51:01.519 --> 00:51:07.920 launch event and will explore the issues in more detail. But UM, I 679 00:51:07.960 --> 00:51:12.760 should say that that, and I mentioned this at the shareholder meeting, that 680 00:51:13.239 --> 00:51:20.880 um Tesla has actually um tremenous capability in real world AI. Yes, in 681 00:51:20.960 --> 00:51:24.880 fact, it is very far ahead of anyone I know. People actually on 682 00:51:24.880 --> 00:51:28.679 Twitter prior to our interview, we're saying, you know, he never gets 683 00:51:28.679 --> 00:51:32.320 asked about how advanced his AI is A Tesla. You always talk about the 684 00:51:32.400 --> 00:51:38.440 other names. Tesla is, like I said, by It's like if there's 685 00:51:38.480 --> 00:51:42.199 not I'm not even sure who's second. Frankly, Um, why is that? 686 00:51:42.199 --> 00:51:44.559 Why? Then? What? What are people not understanding about what you 687 00:51:44.599 --> 00:51:46.320 have? And why are we talking so much about chat, GPT and generative 688 00:51:46.320 --> 00:51:51.239 AI at open AI and what Microsoft can be able to do with it and 689 00:51:51.280 --> 00:51:54.440 not about Tesla. I don't know. I mean people do talk about it 690 00:51:54.480 --> 00:52:00.519 online. UM. I think I think Tesla will have sort of a chat 691 00:52:00.559 --> 00:52:04.840 GPT moment maybe, If not this year, i'd say no later the next 692 00:52:04.880 --> 00:52:07.679 year. UM, wait say that again. I'm sorry you're going to have 693 00:52:07.679 --> 00:52:10.679 a sort of chat GPT moment, Oh you will in terms of suddenly it 694 00:52:10.719 --> 00:52:15.199 will Yeah, suddenly three million cars were a will drive themselves with no one 695 00:52:15.280 --> 00:52:17.159 Right, it goes back to that right yeah, and then five million cars 696 00:52:17.159 --> 00:52:23.159 and then ten million cars. So um, the and and and and I 697 00:52:23.159 --> 00:52:28.239 would I would also say that if if positions were reversed, um, and 698 00:52:28.440 --> 00:52:30.920 say um, well, in fact, the positions are our reverse. For 699 00:52:30.960 --> 00:52:36.159 example, Google has way Moot, which is, you know, sort of 700 00:52:36.719 --> 00:52:38.679 attempting self driving, and they are able to make self driving work in a 701 00:52:38.760 --> 00:52:44.719 very limited geography with with very tightly mapped streets, but as soon as any 702 00:52:44.719 --> 00:52:47.039 as soon as that anything goes wrong with those streets, like there's an accident 703 00:52:47.119 --> 00:52:53.280 or a parade um or road construction, uh, it stops working. Basically, 704 00:52:54.280 --> 00:53:00.880 Google is unable to produce a generalized solution to self driving that works anywhere. 705 00:53:00.360 --> 00:53:02.639 They've been trying to that for a long time. They have been unsuccessful. 706 00:53:02.639 --> 00:53:08.920 Tesla basically has that, and it's far more advanced that than Google. 707 00:53:09.079 --> 00:53:13.199 Yea, And so if the positions were reversed and you said okay, um, 708 00:53:15.519 --> 00:53:20.440 Tesla's got to produce a large language model that has output equal to a 709 00:53:20.519 --> 00:53:24.920 greater than CHATCHVT or Microsoft. Open Eye has to do self driving, and 710 00:53:24.960 --> 00:53:29.719 we just we flipped the tasks. Tesla would win, You'd win. Yes, 711 00:53:30.039 --> 00:53:31.239 you have a computing power and everything else. You'd have to do it 712 00:53:31.440 --> 00:53:36.400 absolutely. Um I'm being told we don't have that much time. Do you 713 00:53:36.440 --> 00:53:38.119 can you give me another five minutes? Or yeah? Yeah, I think 714 00:53:38.159 --> 00:53:44.079 so. I do have a board meeting. I know five five minutes is 715 00:53:44.119 --> 00:53:50.920 probably fine. Sure, okay, thank you? Um um open Ai. 716 00:53:51.079 --> 00:53:52.760 I mean you seem somewhat frustrated with them. You were one of the big 717 00:53:52.800 --> 00:53:59.559 contributors early on the reason I am the reason Opening Eye exists. How much 718 00:53:59.599 --> 00:54:06.480 money did you give them? Um so. Uh, I'm not sure the 719 00:54:06.519 --> 00:54:09.679 exact number, but it's some some number on the order of fifty million dollars. 720 00:54:09.480 --> 00:54:17.480 So the man fate loves irony next level, um So, I used 721 00:54:17.480 --> 00:54:20.480 to be close friends with our page, and I would say at his house, 722 00:54:20.480 --> 00:54:23.519 and we'd have these conversations long long into the evening about AI, and 723 00:54:23.679 --> 00:54:29.199 I would I would be constantly urging him to be careful about the danger of 724 00:54:29.559 --> 00:54:35.519 AI, and and and and he just he was really not concerned about the 725 00:54:35.559 --> 00:54:42.880 danger of AI, and was quite cavalier about it, um and and and 726 00:54:42.880 --> 00:54:45.519 and. At the time, Google, especially after their acquisition of deep Mind, 727 00:54:45.599 --> 00:54:49.599 had three quarters of the world's AI talent. They had obviously a lot 728 00:54:49.639 --> 00:54:52.320 of computers and a lot of money. Um, so it was a unipolar 729 00:54:52.360 --> 00:54:58.360 world for four AI and and go unipolar world. But but the the person 730 00:54:58.480 --> 00:55:01.960 who controls that does not or at least did not seem to be concerned about 731 00:55:02.000 --> 00:55:07.840 AI safety. That that's that sounded like a real problem, so um and 732 00:55:07.880 --> 00:55:15.880 then the final straw was Larry calling me a speciist for being pro human consciousness 733 00:55:15.960 --> 00:55:17.800 instead of machine consciousness. And I'm like, well, yes, I guess 734 00:55:17.840 --> 00:55:22.239 I am. I am a specioust and so right, so you helped to 735 00:55:22.360 --> 00:55:27.159 the creation of open AI. You put as much as fifty million dollars more 736 00:55:27.159 --> 00:55:29.800 than help. It wouldn't exist without it wouldn't exist without you. So I 737 00:55:29.840 --> 00:55:34.400 came up with a name. The name opening Eye refers to open source. 738 00:55:35.039 --> 00:55:37.079 So it's the intent was what's the okay? So what was the opposite? 739 00:55:37.119 --> 00:55:44.000 What's the opposite of of Google? Would would be an open source nonprofit because 740 00:55:44.079 --> 00:55:49.360 Google is closed sourced for profit. Um and that profit motivation can be finicially 741 00:55:49.400 --> 00:55:53.840 dangerous. Um. So uh, should you've gotten governance for that money? 742 00:55:53.840 --> 00:56:01.800 Should you have gotten some level of control? Perhaps in retrospect? Yeah, 743 00:56:01.960 --> 00:56:08.880 I fully admit to being a huge idiot here um so um anyway, So, 744 00:56:08.960 --> 00:56:13.000 so Opening I was like meant to be. Opening I opened as an 745 00:56:13.000 --> 00:56:17.719 open source. It was created as a buy one C three um and but 746 00:56:19.519 --> 00:56:21.880 so so part of it is also in the beginning, I thought, look, 747 00:56:21.960 --> 00:56:24.760 this is this is probably a hopeless endeavor. How could we possibly compete 748 00:56:24.800 --> 00:56:29.800 with How could Opening I possibly compete with with Google Deep Mind? I mean 749 00:56:30.440 --> 00:56:32.880 this, this seemed like an ant against an elephant, you know, which 750 00:56:32.920 --> 00:56:45.639 is not a contest um and um. And I was also h I mean 751 00:56:45.679 --> 00:56:53.440 I I was instrumental in recruiting that the key scientists and engineers, most specifically, 752 00:56:53.800 --> 00:57:00.559 most notably Elias skier Um. In fact, Elio went back and forth 753 00:57:00.599 --> 00:57:04.639 several times because he would say he's going to join an Opening Eye, then 754 00:57:04.679 --> 00:57:07.000 Demis would convince him not to, Then I would convince him to do so, 755 00:57:07.039 --> 00:57:10.440 and then and this went back and forth several times, and ultimately decided 756 00:57:10.440 --> 00:57:15.079 to join an Opening Eye. And really Eliot joining was the was the lynchman 757 00:57:15.159 --> 00:57:20.000 for Opening Eye being ultimately successful. You're very disappointed in what's happened there in 758 00:57:20.119 --> 00:57:28.920 terms of it becoming a for profit. In way, I do think that 759 00:57:28.960 --> 00:57:37.760 there's some look, it does seem weird that something can be UM a nonprofit 760 00:57:38.039 --> 00:57:45.440 open source and somehow transform itself into a for profit closed source. UM. 761 00:57:45.280 --> 00:57:49.239 I mean this would be like like, let's say you funded an organization to 762 00:57:49.280 --> 00:57:53.599 save the Amazon rainforest and instead they became a lumber company and chopped down the 763 00:57:53.599 --> 00:57:58.679 forest and sold it for money. And you'd be therefore like, oh wait 764 00:57:58.719 --> 00:58:00.800 a second, that's the exact opposite of what I gave the money for. 765 00:58:02.119 --> 00:58:07.920 Is that legal? That doesn't seem legal? And if it is, and 766 00:58:07.400 --> 00:58:14.000 in general, if it is legal to start a company as a nonprofit and 767 00:58:14.000 --> 00:58:17.440 then take the IP and transfer it to a for profit that then makes tons 768 00:58:17.440 --> 00:58:25.119 of money, shouldn't everyone start shouldn't that be the default? Um? And 769 00:58:24.920 --> 00:58:30.360 and and then I also think it's important to understand the like like when push 770 00:58:30.360 --> 00:58:35.800 comes to shove, let's say they do create some some digital superintelligence, almost 771 00:58:35.800 --> 00:58:40.199 godlike intelligence. Well, who's in control and what exactly is the relationship between 772 00:58:40.239 --> 00:58:45.079 opening I and Microsoft? And I do worry that Microsoft actually may be more 773 00:58:45.119 --> 00:58:51.320 in control than say the leadership team, and Opening Eye realizes I mean Microsoft 774 00:58:51.360 --> 00:58:54.119 had as part of the Microsoft investment, they have rights to all of the 775 00:58:54.159 --> 00:58:59.760 software, all of the model weights, and everything necessary to run the inference 776 00:59:00.039 --> 00:59:05.159 them. So they essentially have a great deal of control. At any point, 777 00:59:05.239 --> 00:59:09.039 Microsoft could cut off Open the Eye. Elana, I'm being told we 778 00:59:09.119 --> 00:59:13.719 have to wrap up. Your board has been very satient. I want to 779 00:59:13.800 --> 00:59:16.159 end on one AI question. You have a lot of kids. I have 780 00:59:16.199 --> 00:59:21.800 some kids. I have one who's actually soon to go into the workforce. 781 00:59:22.400 --> 00:59:27.400 I struggle with how to advise him about a career when this technology exists and 782 00:59:27.440 --> 00:59:30.400 will only improve. I'm just curious when you think about advising your children on 783 00:59:30.480 --> 00:59:35.320 a career with so much that is changing, what do you tell them there's 784 00:59:35.320 --> 00:59:52.920 going to be a value. Well, that's a tough question to answer, 785 00:59:58.119 --> 01:00:02.480 I guess m I would just say, you know, to to sort of 786 01:00:02.480 --> 01:00:07.679 follow their heart in terms of what they find interesting to do or fulfilling to 787 01:00:07.719 --> 01:00:13.079 do. UM and UM, you know, try to be as useful as 788 01:00:13.119 --> 01:00:19.239 possible to the rest of the society. UM. You know, if we 789 01:00:19.239 --> 01:00:22.119 do get to the sort of like magic Genie situation where UM, you can 790 01:00:22.159 --> 01:00:28.440 ask the A for for anything. Um. And let's say it's even the 791 01:00:28.480 --> 01:00:30.760 benign scenario. Let's say it's a benign scenario. How do we actually find 792 01:00:30.800 --> 01:00:36.320 fulfillment? You know, it's ah, how do we find meaning in life? 793 01:00:36.360 --> 01:00:45.039 If the air I could do your job better than you can? Um. 794 01:00:45.079 --> 01:00:45.880 I mean, if I think about it too hot it correctly, it 795 01:00:45.880 --> 01:00:54.079 can be uh just dispiriting and demotivating. Um. Because I mean I I 796 01:00:54.199 --> 01:00:59.960 go through I mean I I've put a lot of blood, sweat and tears 797 01:01:00.039 --> 01:01:02.480 into building the companies. And then and then I'm like, wait, well 798 01:01:02.920 --> 01:01:07.719 should I be doing this? Because if I'm sacrificing time with friends and family 799 01:01:07.840 --> 01:01:13.400 that I would prefer to do. But then ultimately the AI can do all 800 01:01:13.400 --> 01:01:17.800 these things. Does that make sense? I don't know. Um. To 801 01:01:17.840 --> 01:01:22.599 some extent, I have to have deliberate suspension of disbelief in order to remain 802 01:01:22.679 --> 01:01:27.559 motivated. UM. So I guess I would say, just, you know, 803 01:01:30.039 --> 01:01:35.760 work on things that you find interesting, fulfilling, and that contributes some 804 01:01:35.920 --> 01:01:37.079 good to the rest of society.