Dave Lee
@davelee • 635,897 subscribers
Host of YouTube channel Dave Lee on Investing, TSLA investor since 2012, building with AI. Shift or Noise - new series of 3-minute AI news videos with my take
Videos
After experiencing Cybercab for the past few days in Austin, I think the biggest thing people are missing is the zero-sum game built into Uber. You have two people sharing one environment, with different preferences. The driver likes the temperature at 66. I want 72. If I get more comfortable, he gets less comfortable. He likes his music. I want mine. He wants the windows open. I want them closed. I take a few extra minutes getting to the car, and those are minutes another human spends waiting for me. There’s this constant little tug of war. My comfort and convenience can come at the expense of his. And I don’t blame the driver. It’s his working environment. He’s in that car all day. Of course he has preferences. That’s why I just let him play whatever music he wants. I don’t want my comfort to come at his expense. We’ve gotten so accustomed to this that we barely notice the tension anymore. Then you take a Cybercab. I set the temperature to 72. Nobody gets less comfortable. I play my Spotify at the volume I want. Nobody has to give up their music. I grab my Starbucks and take a couple of minutes to walk over, within the pickup window. Nobody is sitting there getting irritated with me. The car has no competing preferences. That is a profound change in the experience. I can optimize the ride around what makes me comfortable without asking another person to sacrifice what makes them comfortable. And you feel it. You get in, lean back, and relax. There’s no negotiation. No wondering whether your request is going to annoy someone. That whole layer of social tension disappears. I think this is a huge part of Cybercab’s value. Removing that zero-sum dynamic changes what the service feels like at its core. Once people experience that, I think a lot of them are going to have a very hard time going back to Uber.
Dave Lee1,396,704 görüntüleme • 1 ay önce
Trump: “I will end the electric vehicle mandate on day 1, thereby saving the U.S. auto industry from complete obliteration.”
Dave Lee12,973,768 görüntüleme • 2 yıl önce
I attended Tesla’s Cybercab launch in Austin and spent 2.5+ hours questioning engineers about unboxed manufacturing, the plastic exterior body panels, brake-by-wire, the battery pack, AI5 and more. My takeaway: Cybercab may be Tesla’s most important product since the Model 3. I fit as much info as I could in this video. Hope it's helpful!
Dave Lee378,042 görüntüleme • 1 ay önce
I think people are seriously underestimating what Cybercab means for Tesla’s future. For the past few years, everyone has been waiting for a smaller, cheaper Tesla to restart growth. And as that car failed to arrive, the assumption became that Tesla had lost its way or abandoned its high-volume ambitions. I think Tesla is laying the foundation for something much bigger. My read is that Elon made a deliberate decision to give up the obvious next product—a cheaper car with a steering wheel—and concentrate the company’s engineering talent on Cybercab and FSD. That takes conviction. You’re walking away from a product people already want, accepting slower growth, and putting enormous pressure on your teams to deliver. But I think Elon understands something that gets overlooked: a large company can have all the talent and resources in the world and still produce mediocre products because its attention is scattered. Remove the backup plan, and suddenly the entire company has to make this work. That kind of urgency can produce extraordinary innovation. And here’s where I think this gets really interesting. Once Tesla proves Cybercab’s software, manufacturing, and vehicle architecture, it can build an entire lineup on that foundation. Small cars. Larger cars. SUVs. All designed around the passenger. All benefiting from simpler systems, fewer parts, and a fundamentally different way of manufacturing. That’s the path I see back to millions more vehicles a year. Cybercab is the first step in that future. The work going into it today could make the vehicles that follow dramatically better and more affordable. Yes, Tesla is paying a price for this decision right now. But I think the focus it creates gives Tesla a much stronger shot at leading the next era of transportation. Five or ten years from now, I suspect we’ll look back at this period and realize how much of Tesla’s future was being built while people were declaring its growth story over. I break down the reasoning and the path ahead in my new video.
Dave Lee369,420 görüntüleme • 1 ay önce
I created a tennis timing game to help my kids with their actual tennis. Opus 5.5 is amazing.
Dave Lee34,204 görüntüleme • 5 gün önce
Having no steering wheel or pedals in the Cybercab makes a much bigger difference than I expected. In a Waymo or even Model Y robotaxi, you’ve got a steering wheel turning and an empty driver’s seat you can’t sit in. It keeps reminding you that someone/something is missing. In the Cybercab, there’s no reminder that there should be a driver at all. They’ve taken all that stuff away, and you I felt much more relaxed, comfortable and safer. Also, the video-watching (ie., Youtube) experience in the Cybercab is something you just need to experience to understand. For 30 minutes, I almost forgot I was in a car… I think it’s the combination of no driver controls, the huge 22 inch screen, great sound system, spacious interior and being able to lean back and adjust the seats to whatever angle. You start to realize how much of a normal car is designed around the person driving. Once you remove that, the whole experience opens up.
Dave Lee265,004 görüntüleme • 1 ay önce
I asked Fable 5.1 to create an animation about the human body and how people learn. 1,409 lines of code, 87 KB file.
Dave Lee50,180 görüntüleme • 27 gün önce
One side says Tesla will kill Uber. The other says Uber owns the customer relationship and can just aggregate competing robotaxi fleets. I think both miss what changes when you remove the human driver. Uber built an incredible business around a scarce resource: people willing to drive other people around. Recruiting those drivers, coordinating them, and connecting them with riders is difficult. Uber solved a real problem. But a robotaxi network changes that equation. Once you solve autonomous driving safely, you can grow supply by manufacturing vehicles. You don’t have to recruit another human for every additional car. That’s a completely different business. Imagine a company putting enough robotaxis into a city to offer much shorter waits at half the price of Uber. How hard is it to convince someone to download another app at that point? People talk about Uber’s brand, payment relationships, and existing users as if those automatically carry over into a robotaxi world. They matter. But I think people underestimate how quickly a dramatically better service can create demand for its own network. And this is where the “Uber will aggregate everyone else” argument gets difficult. If another company actually solves robotaxis and can deploy them at scale, why would it keep giving Uber a meaningful cut of every ride? Why wouldn’t it build its own network and pass those savings to customers? Uber might help it get started. That doesn’t mean Uber remains essential once it scales. There’s an even bigger problem for the second and third robotaxi networks: they have to compete with whoever gets to scale first. I mean real scale. Enough vehicles, coverage, and availability that the service feels significantly better to the customer. A few impressive demos don’t establish that. Once a network reaches that point, the next competitor has to match its availability AND its price. That could require a lot of subsidies while building out a fleet. This is why I think Tesla’s work on Cybercab manufacturing is such a big deal. A competitor needs great autonomous driving software. But it also needs a vehicle cost structure that lets it compete sustainably. Tesla is working on both sides of that equation. If Tesla gets the software right and scales a much cheaper vehicle, that combination becomes extremely difficult to compete against. Does that mean Uber goes bankrupt overnight? I don’t think so. There are plenty of services where people want or need a human involved. Delivery, assistance, specialized transportation. Habits also change more slowly than people in the tech world sometimes assume. Uber has room to adapt. But I think a true robotaxi network becomes a different class of transportation: cheaper, more available, and useful for far more trips. The big questions are when Tesla can reach the required safety and reliability, how quickly it can deploy, and who else can match the complete service at a competitive cost.
Dave Lee90,547 görüntüleme • 1 ay önce
Day 3. Finished 5 Tesla Robotaxi rides, 4 Waymo rides, and 1 Uber ride. (See 39-minute edited video below.) Uber is in a tough situation. Classic innovators dilemma. Autonomous ride-hailing clearly is the future, but if Uber invites Waymo then that's working against their human drivers which will hurt Uber's recruiting ability (which they need to keep density especially in suburban areas). But if they don't welcome Waymo, then Uber is toast because people want cheaper/autonomous ride-hailing. Also, if Uber welcomes Waymo and Waymo cars increase then Waymo can just bypass Uber with their own Waymo app. Uber's only "hope" would be if multiple autonomous driving companies come online soon and Uber becomes the ride-hailing marketplace for them. But I think this is unlikely. Autonomous driving is just a really hard problem to solve, and with data and network effects. I think it's a winner-takes-most market. We'll see. But the more I take Tesla's Robotaxi, the more I think it's smoother and better than Waymo. And the more I take Waymo, the more I get pessimistic about Uber's long-term future.
Dave Lee593,539 görüntüleme • 1 yıl önce
This afternoon I picked up a new Nvidia DGX Spark computer with the goal of trying to run Gemma 4 31b (4bit) on it locally as a server. Just 1.5 hours later, it’s working! Using Open WebUI on my MacBook as the interface and it’s connecting to my DGX Spark running as a Gemma 4 server.
Dave Lee141,105 görüntüleme • 6 ay önce
This year has been a trying year for my family. Wife has had to deal with surgery, chemo, and now radiation for stage 3 breast cancer. To help my kids cope and thrive I’m having my kids hit a lot of balls everyday (table tennis and tennis). It’s part distraction but it’s also a way for them to grow in their coordination, reflexes, athleticism, and more. It keeps them positive, active and eager to learn and grow. Focus on the ball and hit it. Hundreds of times everyday. Something good will come out of it. At least that’s my hope.
Dave Lee240,697 görüntüleme • 1 yıl önce
