Big Tech lands an early win in legal battles against publishers

This week, two major AI companies scored early wins in court, with federal judges siding with Meta and Anthropic in separate lawsuits over how their models were trained on copyrighted material. The decisions represent the first real legal validation of AI companies’ argument that training models on books, images, and other creative works can be considered “fair use” — even if those materials weren’t obtained with permission. It’s a big deal for companies building generative AI, and a potential turning point for the many lawsuits still in motion. Today, on TechCrunch’s Equity podcast, hosts Max Zeff and Anthony Ha were joined by Sean O’Kane (who graciously stepped in while Kirsten headed off to the Nevada desert to see the next big act of Redwood Materials, the battery recycling and materials startup founded by former Tesla CTO JB Straubel) to dive deeper into the rulings. While neither case sets a precedent yet, Anthony noted that appeals are likely, and broader challenges could ultimately shape how AI companies interact with entire industries going forward. Listen to the full episode to hear more highlights from the week, including: Kalshi’s $185M round, and what it says about the rising (and legally murky) world of prediction markets The startup betting on reusable satellites, and why the Department of Defense is paying attention Tesla’s robotaxi rollout in Austin, and how it stacks up against Waymo and other AV companies’ approaches Equity will be back next week, so stay tuned! Equity is TechCrunch’s flagship podcast, produced by Theresa Loconsolo, and posts every Wednesday and Friday. Subscribe to us on Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod. Credits: Equity is produced by Theresa Loconsolo with editing by Kell. We’d also like to thank TechCrunch’s audience development team. Thank you so much for listening, and we'll talk to you next time. Learn more about your ad choices. Visit megaphone.fm/adchoices

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Anthony Ha?

  This episode of Equity is brought to you by Schneider Electric. Schneider Electric, advancing energy tech.

• 0:17Speaker 01

  Anthony, do you ever just want to start off an interview really combative?

• 0:20Anthony Ha?

  I think that's kind of my default state. I just like to like really get people off their guard and destabilize them. It sounds like that's something you actually saw in the last few days.

• 0:30Speaker 01

  Yeah, I mean, respectfully, Anthony, you're a nice guy. I can't imagine you doing that. Sam Altman, however, he came on stage in San Francisco at an event I was at this week, swinging. I think is the only fair way to put it. It was an interview with the New York Times, the Hard Fork Podcast's Kevin Rus and Casey Newton. And Altman barged on stage too early, like while they were like introing him and teeing him up. And there was this like awkward moment and then he immediately started asking them questions about why the New York Times was suing them and asking for their personal opinion, which was, you know, it's fun just to be, you know, combative right off the start.

• 1:10Anthony Ha?

  I mean, I do feel for them because definitely if I was on stage with someone that TechCrunch was in litigation with and they started asking me questions about that, that would be pretty, pretty tough, I think. It's also funny because anyone who reads the New York Times' coverage of, I think, AI in general, but definitely of OpenAI, like you just, every story they write, they rightly have to put at some point parenthetically, "The New York Times is suing OpenAI over copyright issues." But it's there so often that now it just feels completely invisible to me. And that's just like part of the furniture when they write about it. But clearly it is not invisible to Sam Altman.

• 1:47Speaker 01

  Oh, no, I think, I think he thinks about it quite a bit. And we're going to get into more of those copyright issues that are coming up with the tech industry. It's in a really interesting spot right now. We'll talk about that later on the show. But welcome back to Equity, TechCrunch's flagship podcast about the business of startups. Today is Friday, June 27, and I'm Max Zeph, a senior AI reporter with TechCrunch.

• 2:09Anthony Ha?

  And I'm Anthony. TechCrunch's Weekend Editor.

• 2:11Sean O'Kane?

  I'm Sean McCain, a transportation reporter, filling in while Kirsten gets to do the fun stuff, driving a mysterious car that we can't talk about.

• 2:19Speaker 01

  Ooh, mysterious.

• 2:20Anthony Ha?

  But you know what it is, Sean.

• 2:22Sean O'Kane?

  I am sworn to secrecy. I cannot confirm nor deny.

• 2:25Speaker 01

  Wow, I'm on the edge of my seat. But on the show today, we have Deals of the Week from Calci and Lux Eterna, and then we'll get into the growing tension around AI copyright lawsuits, plus an update on Tesla's robo-taxi expansion and what it means for autonomy on the ground. Let's get into it with the first deal. Calci, the prediction market, is locking in a $185 million round as its decentralized rival, Polymarket, reportedly eyes an even bigger raise.

• 2:54Anthony Ha?

  So Calci, I should say, is a prediction market that I haven't used myself, but it has underlying blockchain technology. I don't know. You can't see me, but I'm sort of like waving my hand here. But it's one of the sort of well-known prediction markets. It just raised $185 million at a $2 billion post-money valuation. I mean, first of all, I'm just curious, have Max or Sean, have either of you used Calci?

• 3:17Speaker 01

  Yeah, I tested them out a little bit. I actually interviewed Calci's CEO a while back, so I played with it a little while then. And just in the election for New York's mayor, I was playing with it a little bit because there was a lot of betting going on there.

• 3:32Sean O'Kane?

  I've not used either, but I've always said that gambling, the only thing it was missing was the blockchain.

• 3:40Anthony Ha?

  Yeah, I mean, I honestly, like, because I've definitely heard of Calci before, and I live in New York and was definitely like following the election very closely. And I think that, you know, I don't know if it was the Calci stats that they were showing, but like, the people were sort of making a big deal about that in the betting markets, like Zorn Mom Donnie, who did ultimately win the primary, like, kind of the last minute over to Andrew Cuomo. So like, the prediction markets were kind of at the last minute. There was some swing in the correct direction. But it just seems like, in general, there's just been this huge uptick in the kinds of things that you can bet on. Although, the legal structure on that is also a little bit uncertain. And, you know, Polymarket in particular is not operating in the US and has been banned since 2022.

• 4:24Speaker 01

  Yeah, I think that there is kind of an interesting thing going on with these prediction markets. And I remember when I talked to Calci's CEO, he was telling me that, you know, they are an information platform, so to speak, and not just a gambling site. And he made a lot of arguments that felt like they were curated by his lawyers about why they are not a gambling site and that they were providing information to people. And I, at the time, I was pretty skeptical of that idea. You know, I thought that was just kind of a fancy way of, you know, getting regulated by federal regulators instead of state regulators to avoid, you know, taxation and a bunch of state laws. But you did see

• 5:07Speaker 01

  with the New York mayoral race that, you know, Calci and Polly Market did have kind of the results of, it was the primary race, but they did have the results of that before a lot of news outlets did. And it was interesting to see people pointing to it and say, look, it's over. And it's like, okay, is that true? I don't know. But that's what people are pointing to, you know?

• 5:28Anthony Ha?

  Yeah. And I think you're right to point out this, it's very convenient for the CEO of Calci to say like, this is an information platform. But obviously, they are driving this tremendous amount of conversation. I think part of the question is like, can they keep that going beyond things like the presidential election, the mayoral election? And I think that when we think about it, not just as a gambling platform, there's also a question of like, a strong argument, and frankly, an argument that I agree with, right, that we over obsess, certainly, American media over obsesses with sort of the horse race aspect of a lot of our politics. And so like, there's more than one relevant question about the mayoral primaries, not just who's going to win and who knows it first. It's also like, who should win? Who aligns with your values? And when we fixate so much on who's going to win, that sort of gets obscured sometimes.

• 6:13Sean O'Kane?

  I feel like there's two things that jumped out to me about this and just these betting markets in general. One is that this part, I think, goes unspoken a lot, is that there's just kind of an uptick in gambling addiction, probably speaking in this, in this country. And I was haunted by this video all week on kind of every social media platform of like, two guys betting $100 on whether the gambling hotline operator would be a woman or a man, you know?

• 6:39Anthony Ha?

  Oh, my God.

• 6:40Sean O'Kane?

  It's its own content in a way, which is like, I think, half the deal with these, with Polly Market and Calci, where like,

• 6:46Speaker 01

  Not only are they trying to produce quote unquote information and obviously enabling betting, but they're also like their own form of like entertainment. And the other thing that comes to mind is there's like a through line between this and what we'll talk about later with the Tesla robot taxi, which is like, in my view, if you don't pay the closest attention to what they're doing, they look like they work, but only up until they don't. And you know, like they look like they're really predicting what the true outcome is up until the moments where they don't and they get it wrong. And you see those like very last minute reversals that almost neatly track with the actual news breaking.

• 7:22Anthony Ha?

  Right. And I mean, in the case of Tesla, obviously the failures can be a little bit more spectacular or like more

• 7:27Speaker 01

  Sure.

• 7:28Anthony Ha?

  obviously dangerous, but like the, you know, the growth of sort of gambling in general betting markets, prediction markets does also have like real world consequences. I mean, I think John Oliver did like a great segment about this. So it just, I mean, there's a whole range of like harmful behavior and it's, it's not just the people who are like the hardcore gambling addicts, but it's really bad for like the people who get super addicted.

• 7:47Speaker 01

  No, I mean, I don't think that there is a convincing argument to me that this is not just gambling. Like I, I think it just is despite what the company says. And I think, you know, the information point of it is it's not like this is a source of information. People are getting information from other places and then this plays out on Kelsey with their bets, but it's not like, you know, this is where people are getting it from exactly. You know, like, I don't know, like that to me feels like an important distinction that just is like, you can just see it, but it's not where people get it from.

• 8:19Anthony Ha?

  So. And it's also a place where, and again, I don't think this is going to happen super often, or at least I hope not, but I think there was definitely a lot of discussion around the betting markets in the presidential election and like that if you have a lot of money to throw around and you have a favorite candidate, you can swing the prediction markets a little bit in the direction that you want. And something's like, oh, like, you know, Trump is doing better than we expected or Harris is doing better than we expected. And so I myself fall victim to thinking it's interesting and fun to talk about and paying attention to like what the prediction markets are saying, but I think we should all be like pretty careful about the extent to which we do that.

• 8:52Speaker 01

  Yeah. I mean, nevertheless, these are getting like massive rounds right now. So it's hard to ignore that these, just the amount of money that's flowing through these. And I think the potential that investors see for these to become just so much more widespread. And you can kind of only imagine what people will be betting on in the future. So that said, we need to move on to our next deal, which is from Luxeterna, which is a story Sean had. Sean, what's going on here? And also, how do you say that name?

• 9:21Sean O'Kane?

  I think you can choose your own adventure here. I don't think they're trying to be too offbeat with the name. I've just been calling it Luxeterna. Yeah, this is a, this is an interesting one. It's obviously a much smaller funding round than what's going on in the sort of prediction markets. But this is a new Denver-based startup founded by, in particular, the CEO. He used to work at SpaceX. He worked at Amazon's Project Kuiper satellite program. He also worked at Loft Orbital, which is one of the buzzer sort of space launch and integration startups over the last few years. You know, I find them really interesting because they, they raised 4 million in pre-seed funding led by Space Capital, Dynamo Ventures, Mission One Capital were involved in that and a few others. And their whole thing is just building reusable satellites.

• 10:04Sean O'Kane?

  And you know, for people who don't watch this space that closely, that might not seem like something that is all that out of the ordinary, but it actually still kind of is. China's been working on it a little bit. There are a couple of other startups that have sort of touched on similar ideas. But this is one of the first real sort of headlong attempts to make a satellite that can go into space. Survival reentry, which is really challenging, and then go back up and do that again and again. The ultimate goal, at least in the initial run of Luxeterna, the CEO told me is, is to make something that could withstand 15 launches and reentries, which is, you know, would be pretty impressive. Their big thing is this is possible now because the cost to get to space is so cheap,

• 10:47Sean O'Kane?

  in large part thanks to SpaceX, and because SpaceX and Blue Origin are developing these massive heavy lift rockets, Starship in SpaceX's corner and New Glenn in Blue Origin's side, that can just frankly fit larger things. And the more space you have to play with as far as what's being launched into orbit, the more options you have for design. You can think about it this way, up until, you know, pretty recently. Basically, any mission, science, commercial, whatever it is, if you were making a satellite or a spacecraft, you're always starting with what's the maximum size I can fit in that nose cone or that cargo space and working backwards from that constraint, almost more

• 11:28Sean O'Kane?

  so than than money. And so just having more size opens up new options. And the big thing is that he wouldn't go into the specifics of it, but the big thing seems to be that, you know, they have this kind of conical heat shield. It makes their satellite design look a little bit like these kind of classic NASA spacecraft. And it seems like the satellite structure itself will sort of fold in or tuck in behind that heat shield so that the heat shield's taking the brunt of the force when it's coming back into Earth's atmosphere and protecting all of the components and stuff behind it.

• 11:58Anthony Ha?

  So you mentioned, you know, their tech, I guess, is sort of at least inspired. I don't know how directly taken from this NASA design. But as I understand it, like, you know, they haven't had actually a satellite go up and survive reentry yet. Did you get a sense of like how confident he was that the first one will be successful? Or are they seeing it as much more of a test and it can kind of go either way?

• 12:18Sean O'Kane?

  I would say it's probably leaning, if at all, maybe a little bit towards the second option there, if only because spaceflight can be so unpredictable. You can model this stuff out to the nth degree and, you know, you really just have to test this stuff out to know for sure. They haven't launched anything yet. The first planned launch of a sort of smaller scale version of the satellite design is going to go up in 2027. They already have the spot booked on a SpaceX Falcon 9 rocket. So that's the target they're working for. They're a small crew. They only just started seven months ago. There's only about seven of them on staff. And Brian Taylor, the CEO, told me, you're probably only going to grow to about 20 or 25 people by that 2027 launch. So they're they're doing this

• 13:01Sean O'Kane?

  tight and very light and and just focusing on something very specific and trying to get that 2027 launch to work as well as they can. And then from there, they'll maybe grow a little bit more and and maybe by then they'll have a little bit more funding to play with.

• 13:16Anthony Ha?

  I mean, obviously, a reusable satellite, presumably that like saves money. Are there other benefits? Or do you think that changes potentially like what satellites are used for?

• 13:26Sean O'Kane?

  It definitely will cut down on costs. The other really big thing is, right now, you put a satellite up into orbit, it's up there. I mean, five to 10 years, in some cases forever. Operationally, five to 10 years, kind of regardless of what happens. Usually satellites either deorbit and burn up in the atmosphere, or some of them actually get pushed up to safer orbits away from other space debris. But that makes things really inflexible. It means that you have to spend all of this money, hundreds of millions of dollars sometimes. Or if you're a smaller company or a science outfit that's just trying to do research, you know, you have to tack yourself on to these larger satellites. And you have one shot at it. It goes up and it's up there. If something's wrong, you can't go fix it, basically.

• 14:08Sean O'Kane?

  And so just the ability to access the satellite again, bring it down, maybe put new cameras on it that have better capabilities than just two or three years ago. There's just so much more flexibility in play. And it opens up a lot of different options that modern satellites don't have, even the most advanced ones. You know, I think a lot of people now think satellites, they think of SpaceX as Starlink. And they've come a long way with that design in just a few short years. But those things are still basically just built to last like two or three or four years. And they come, you know, a lot of them just deorbit even earlier than that and burn up. And so SpaceX has been in this kind of race of like outpacing the ones that deorbit. So having that flexibility is a huge thing that really changes a lot. And I think the big thing that Brian Taylor told me when I talked to him was,

• 14:52Sean O'Kane?

  you know, this feels very much like an immature industry still. And that the ideas of what could be possible can only be realized when you have technologies like this at play. And so it can open up a lot of really exciting stuff.

• 15:04Anthony Ha?

  One thing that I've read about recently that I don't think this is a luck to turn a problem, but it, I guess, gave me an opportunity to mention it was also just that as this opens up, we're starting to see like, you know, the Starlink satellites, other kind of like nano satellites potentially going up is we're at the very beginning of seeing more issues around sort of light pollution and things like that, too, is now that like you look at the night sky and suddenly it's, yeah, I think littered with, is it fair, littered with all these other manmade objects. And that seems like something that's also just going to increase. And that seems like also an area where establishing some more like norms around what is and isn't okay is going to be important, too.

• 15:36Sean O'Kane?

  Yeah, you know, I think with this, they're not looking at doing crazy volume anytime soon. There's a lot of run up to when this could actually be happening. At any kind of scale, if anything, the promise of reusable satellites means there could theoretically be not as many vehicles in space. And if anything, you know, helps a little bit in the long run with the space debris problem. I mean, a lot of that, you know, we'll see. Maybe it turns into a situation like induced demand here on Earth where, you know, you, you build a wider highway thinking that it's going to free up traffic and it just creates more traffic. So maybe it'll, it'll break the wrong way. But you know, their whole, their whole business is really going to be built around essentially being something that you can kind of plug into. You know, they're not trying to operate their own missions or anything

• 16:21Sean O'Kane?

  like that. They will get the satellites up there on these other rockets, but they want to be a sort of platform really for other people to launch again, sort of science commercial, but it really defense. I mean, the Department of Defense have been watching what they're up to really closely. They're already working a little bit NASA. And, you know, there's just a lot of money coming into not just the defense space right now, but the sort of orbital defense sphere. And so I think this could be a popular way for those kinds of government agencies to get up into orbit cheaply and again, flexibly.

• 16:55Speaker 01

  Okay, that's fascinating stuff, but we do have to take a quick break. As promised, we are going to get right into the AI copyright battle after this.

• 17:08Anthony Ha?

  What moves our world forward? Energy. History has been shaped by it. Revolutions won with it and progress dependent on it. For 190 years, Schneider Electric has been a leading force in this story, creating the technologies that harness energy and transform the competitive landscape. And today, in the dawning era of energy and industrial intelligence, Schneider is helping customers build more efficient, resilient and sustainable operations across every sector, from data centers to homes, from buildings to industry. Schneider Electric, your energy technology partner. Discover more at se.com.

• 17:44Anthony Ha?

  All right. So Max, it seems like there's been this kind of rapid succession of all these different rulings in a number of different copyright cases around AI related to, but not actually the case that Sam Altman was complaining about this week. So can you give us just kind of a quick rundown of what's happened?

• 18:00Speaker 01

  Yeah, there's been a lot of activity here. So earlier this week, a federal judge sided with Meta in its lawsuit over training AI models on copyrighted books. The judge found that Meta's training on copyrighted books was fair use, which means that it did not infringe on the copyright protections of these 13 authors. And then there was another case in which Anthropic also won its lawsuit over similar claims that had been brought against them. Basically, book authors had alleged that Meta had infringed on their copyright by training AI models on their stuff. The judge said, this doesn't really infringe on your copyright. And what's interesting about these cases is that they're potentially

• 18:44Speaker 01

  a very big deal because I think a lot of media creators and publishers have filed these lawsuits in recent years to try to get AI companies to fairly use their data and compensate them for using their data to build these huge businesses that AI models are becoming. And I think these are the first indications that maybe copyright holders and publishers don't have as much of a claim to their work in this context as a lot of people hoped for, and a lot of people thought they did. I think there's a couple important caveats here, but yeah, I think we can just start there, that this is kind of a big win for the tech industry on this front.

• 19:27Anthony Ha?

  Right. I think I first heard about these rulings on one of my days off when I was just browsing Blue Sky, which is not generally a super pro AI environment. And I think there was a lot of alarm about these rulings. And I don't think those are unfounded, but as I dug into it more, it does seem like, yes, these are definitely victories for Meta and Anthropic specifically, and maybe the tech industry more generally, but that they're also designed to be relatively narrow, that neither judge is trying to say this is what copyright and IP law around AI training looks like going forward. It's just saying, you don't quite have the case. There is some fair use defense here.

• 20:12Speaker 01

  Yeah, I mean, I think that in the Meta case specifically, there was an interesting argument that the judge made, which is basically that there was basically no question that Meta did train on these books from these 13 authors. It was like, that was very clear. And Meta did not pay these authors for it, and that much was established in the case. But the real thing that kind of broke this case was that Meta did not harm the market for these books, which is what the judge kind of had established. And really, it was the plaintiffs, did it make a sufficient argument that, you know, there was less of a market for their books in light of this happening? And that's a really key part of copyright law. It's not just as like boilerplate as like, hey, that's mine, you can't use it.

• 20:55Speaker 01

  It's like, does this really harm my ability to make a living doing this kind of stuff? So that was kind of an interesting part here. And then the Anthropic case, I mean, there was a whole kind of section about Anthropic's pirating millions of books. I mean, Anthropic went to great lengths to pirate books online. When they realized that they might be doing something wrong, it actually came out that Anthropic was buying like used books and ripping the pages off and scanning them to put them into AI models. It was this like huge operation that they ran. And this is just like insane to me that like, these are the lengths that these companies are going to to get this data. But yeah, but it still might not be fair for the authors to get paid for it.

• 21:36Anthony Ha?

  Well, it seemed like the crux of, I think it was the Meta one at least, was this idea that you have to have, I guess this is also coming up, this is something that's going to be sort of discussed in the Anthropic case as well as this question of like, did they obtain the books legally? Like, if you didn't obtain the books legally. But then, I mean, it seems to me like that's a pretty low threshold. If I'm understanding properly, like, there's literally like, Mark Zuckerberg can just hire somebody to go and buy a copy of every single book at Barnes and Noble or Amazon. And then now he's obtained them legally, and that's all he has to do to train on those books?

• 22:07Speaker 01

  It's complicated. I mean, like, I think that the thing that these cases have highlighted is that, you know, if tech companies pirate books, that is illegal. Like, just because you put them in an AI model doesn't make it legal to, you know, pirate millions of books. But, but then, yeah, if you buy them legally, it still also could be fine to train AI models on it. I think the key things that the judge has said here was that this is, applies specifically to books. This does not apply to news articles. This does not apply to films and TV shows. This doesn't apply to music. And I think the argument that they're making is that AI models really might harm the market for those other things. Like, I think there's a legitimate argument to be made that AI is going to hurt the market for news articles.

• 22:49Anthony Ha?

  I think there's a legitimate argument to be made that it already has.

• 22:52Speaker 01

  Yeah, right, right. So, I mean, I think that we'll have to wait and see, you know, what happens in the rest of these cases. And to me, this has clarified that we're not going to get a big sweeping ruling that says AI companies can't use all content, you know, at all. They have to pay for all of it. It's like, it's probably going to be an industry by industry basis now, which is really fascinating.

• 23:15Anthony Ha?

  Yeah, because these are also like individual judges. So I would imagine that probably these are cases, however they turn out, they're probably going to be appealed and taken to higher courts as well.

• 23:24Speaker 01

  It seems likely. It seems likely. It's unclear to me which one will kind of become the law of the land here. It doesn't seem to me that either of these cases are like setting a broad rule here, but we will have to see, you know, obviously, the New York Times lawsuit against OpenAI is another closely watched one here. That one definitely feels like it could set a precedent for a much broader set of content. So we'll have to see. With that, we are going to kind of change gears a little bit here to talk about some unanswered ethical questions. I think I have ethical questions about this. I'm not sure if that's the whole story here, about Tesla's robo taxi expansion. Sean, can you explain what happened this weekend with Tesla's robo taxis?

• 24:10Anthony Ha?

  Sure. How much time do you have?

• 24:11Sean O'Kane?

  Uh, no. So we, you know, for years Elon Musk has been promising that Tesla's cars would be able to drive themselves. This promise has taken a few different forms. He's promised, you know, most famously, I think in 2016 that, you know, every car that was being built from that point forward had the hardware necessary for Tesla to just be able to flip a switch and make those cars self-driving. That has never come true. They have gotten increasingly close to that with the autopilot software that they started with. And then the, you know, what they call the full self-driving software that has been on people's cars for the last few years. The other big promise that he was making, and this especially was promised in 2019, was that the company would basically make an entire robo taxi fleet.

• 24:56Sean O'Kane?

  He had said at that time that there would be a million robo taxis on the road in 2020. That clearly didn't happen either. And so now we're at this point in time where there's sort of two things happening at the same time. One is that Tesla in October last year revealed this thing called the Cybercab, which is basically a purpose-built quote unquote, robo taxi that doesn't have a steering wheel, it doesn't have pedals, it only seats two people. And that is a thing that they see as like the future of where this is all going. But that's not in production yet and won't be for a little while. So in the meantime, they still think that their software is good enough to operate something like what Waymo has been doing with a real commercial robo taxi fleet with just the Model Ys that exist today, the Model Y SUVs. And so they finally

• 25:40Sean O'Kane?

  launched a very limited version of that product in Austin, Texas, starting on Sunday. And it's only open to people who are invited to the program. It is a reduced version of the vision that he's been promising for many years. And we still have many questions that aren't being answered by the city, especially by Tesla. And we've even tried some other methods. We submitted a records request to the Department of Transportation in Texas. Tesla tried to block that request. Can fortunately say that it appears the Attorney General is not siding with Tesla on most of that, according to a letter I got the other day. So we'll see what we get from all that. But, you know, the, you know, it is, it is similar to what you would see from a Waymo in the cities that it's operating.

• 26:21Sean O'Kane?

  And there's a dedicated app. You pull it up, you hail a car, it comes to you, it picks you up and takes you to your destination. The difference here is that Tesla's still got a quote unquote safety monitor in the passenger seat. There's nobody in the driver's seat, but that safety monitor does have the ability to step in and tap something on the screen to, you know, stop the car, pull it over. There are some questions about whether or how many people are remotely monitoring these vehicles from Tesla headquarters in Austin, or maybe even other places like their offices in the Bay Area. And, you know, it looks like it has gone more smoothly than most people really thought it would.

• 27:01Anthony Ha?

  But not a hundred percent.

• 27:03Sean O'Kane?

  But not a hundred percent for sure. There have definitely been a few moments where it's messed up. And you think about how small this fleet is at the outset. It's really only about 11 vehicles so far that the people have been able to locate. To have a few dozen rides happening over the first few days across 11 vehicles and still have a couple of those kind of scary moments that people see on their regular cars using full self-driving around the country and around the world. You know, it seems to imply that maybe Tesla hasn't quite gotten the technology there to really sort of rip the Band-Aid and turn this into like, you know, something that we would legitimately consider a robotaxi service. So

• 27:39Sean O'Kane?

  that's kind of where we're at, you know? There had been some reporting that they were going to do this earlier in June. Musk said that they wanted to make sure they were doing this as safely as possible. So they pushed it back, but you know, in my view, there was no way they were going to put this off any further because next week we find out how many cars the company delivered in the second quarter. That's not expected to be very, a very good number. And then, you know, we find out their financial results for the quarter later in July. So, you know, to me, it was clear they were going to try and do something this week to get some good news out there.

• 28:06Speaker 01

  Yeah. I mean, Sean, it, it seems to me like, you know, when you look at Waymo is doing right now, that they are just above and beyond like what, you know, Tesla is really doing with their robotaxis. You know, the fact that Tesla needs someone to sit in the car with you while you get in the robotaxi feels incredibly telling to me. And what do you make of, you know, Tesla's spot in the robotaxi race right now? And do you feel like you can, you should count them out based on this? Or do you feel like a year from now, we could be in a different place?

• 28:38Sean O'Kane?

  I think we just need a lot more information, you know, like this is so limited. It's so hard to know exactly what's going on. Some of those like kind of scary moments that we saw this week. Were buried in 20 minute videos from these influencers who were less likely than most people to point out a moment that was kind of scary, right? So like, we just, there's a lot more we need to know. I do think one piece of context that's important to consider in all this is that Waymo didn't just snap its fingers and put a bunch of robotaxis on the road and start offering 250,000 rides a week, which is what they're up to now. They went through these steps that Tesla is kind of mimicking to a certain extent. They had safety drivers, they were manually driving in cities, they still do that in new cities before they launch.

• 29:22Sean O'Kane?

  Then they switched to safety drivers, kind of monitoring the system, ready to take over if anything goes wrong. And they slowly moved to the point where it wasn't just immediately you're in a Waymo and there's nobody driving at all or nobody sitting in the car at all. So they progressed there, Tesla's trying to do that, as is the case with pretty much anything Elon Musk does, he's clearly trying to speedrun that process so that he can get ahead of where Waymo is at now. And so I think we just need to learn more information about how this is really going, what it looks like when it scales up beyond just like 11 vehicles and you have people who are genuinely using the system for transit versus, I mean, most of these people who were in testing the service this week don't even live in Austin. I mean, they were flying in from California or the East Coast or wherever.

• 30:05Sean O'Kane?

  Because they got, and it's what Tesla does, these are the people who show up at their shareholder meetings, these are the people who show up at all their events, they don't really invite press to anything anymore. And so it's hard to get a real sense of how it's been going. In that regard, it's been a real Rorschach test. I mean, this week has been exactly what people who hate Tesla needed to say, this is super dangerous and it's going to kill everybody on the road. It's also been exactly what all the fans and biggest supporters have needed to say. Tesla's done it, Uber's dead, Lyft is dead, Waymo's dead, Elon Musk wins again. And we're not at any point where we can say anything definitive about any of that stuff.

• 30:44Anthony Ha?

  Yeah, I mean, two things that maybe are worth like mentioning or underlining is that in terms of like the not 100% successful side of it, it's not just sort of like social media footage, I guess, the National Highway Traffic Safety Administration has also kind of reached out to ask that. I mean, I guess they are basing that on the video, so they don't necessarily have more information than we do. But there's some action. I mean, we'll see. I mean, I also wonder, you know, how, you know, the constantly evolving relationship between Donald Trump and Elon Musk will shape that. And then my sense is also that for Tesla, you know, A, like, there is a sense of, you know, maybe robo-taxis are going to be a big business in the future, and they want to be in that, but also that self-driving more broadly is really important for Tesla because they are treated by Wall Street as a technology company.

• 31:28Anthony Ha?

  And particularly, it's like, it seems like the EV market is, there's a lot of uncertainty there. They have to be able to convincingly make the case that we have this breakthrough technology that no one else does. Otherwise, you cannot justify the, you know, share price that their stock is trading at.

• 31:42Sean O'Kane?

  Yeah, this has been the thing that's been driving the stock price for years, is this promise. It's one of the reasons that Tesla has been under so much scrutiny around these claims is that it's not just, you know, they were aspirationally trying to do this. They were saying for years that this was coming, and it never came. And that the belief that it was around the corner is why the company is valued so high. And I should say, something that I think has been really overlooked this week is that those promises have changed a little bit very quickly. The original promise of what Tesla was doing was that they were developing an end-to-end, you know, AI-driven, self-driving software that you could put anywhere, and it would go anywhere. And now they're doing something that looks more closely resembles what Waymo and other autonomous vehicle companies do, which is setting a geofence,

• 32:27Sean O'Kane?

  mapping and gathering data of that area ahead of the launch so that the cars have a little bit of an edge, knowing what they're about to encounter. That was a big concession for Tesla. And the other thing is that that original promise from 2016, that all these cars that were being made have the hardware necessary to do this, and that there would just be this flip-the-switch moment. That's no longer the case. Elon Musk even recently, a few months ago, admitted that many, I mean, we're talking like a million or more of the cars that are on the road now that have older versions of that hardware that were built between 2016 and now are not up to snuff to be able to power this software.

• 33:09Sean O'Kane?

  And you look at the language that they've used this week, especially now they're saying every car that we build today can do this, which is a big change from 2016. All these cars are going to be able to do this. And so I think the closer we get to this being more of a real launch and that we get to that real moment where they try to sell that vision of your car can do this too, that we're this thing that we're doing in Austin, that will, I think, trigger that moment where we'll see a bunch of legal fights over, hey, you advertised to me, this is why I bought this car years ago. You told me this was going to be an appreciating asset. It was going to go out in the world and make money for me. So I think there's a lot of reckoning to come on that side of things.

• 33:53Sean O'Kane?

  And I think it's been really overlooked this week because, you know, that was a really wild admission for Musk to make. A lot of people theorized that that was going to be a problem for years. And for him to cop to it earlier this year was a pretty big moment that I think has been overshadowed by what's happening in Texas.

• 34:07Speaker 01

  Yeah, I mean, this is a real make or break moment for Tesla. And it'll be interesting to see how this really plays out in the coming weeks. That said, we are just about out of time. Equity will be back next week. But until then, you can find Equity under the handle at EquityPod on X and Reddit. We'll see you next time.

• 34:28Rebecca Bellan?

  Equity is produced by Teresa Locan Solo with editing by Kel. We'd also like to thank TechCrunch's audience development team. Thank you so much for listening and we'll talk to you next time.