저자: DanIves · 외국인
요약 보기
Dan Ives는 2026년 테슬라 목표가를 베이스 600달러, 불 800달러로 제시하며 로보택시 30개 도시 확장과 데이터 우위를 근거로 든다. 이어 Boston Dynamics는 데모 중심이고 현대차 공장 실사용·경제성 면에서 테슬라 옵티머스가 더 실용적이라는 해설이 나온다. 마지막으로 Dan Ives의 2026년 top5로 테슬라, 마이크로소프트, 애플, 팔란티어, 크라우드스트라이크가 언급된다. 엔비디아도 1순위로 언급된다. 자막 후반부는 광고성 멘트가 섞여 있다.
다듬은 글
I've got three spicy clips for you today. Firstly, Dan Ives just laid out why 2026 could be a huge year for Tesla. Then in clip two, we've got the CEO of Boston Dynamics talking about humanoid robots doing real factory work. And I think there's a connection here between those two stories that's worth exploring.
And finally, we're going to see Dan I's top five stock picks for 2026. Enjoy. Tesla 600. Um, it's not a car story. Do you see 800? Do you see a thousand? Would that be crazy or no? I mean, that's where it's headed. Base case 600, bullcase 800. I mean, to me, when it comes to Tesla, that's one like this is the most important year I'd say for Musk maybe ever because of autonomous, because of robotics, because of where they're heading down this AI path.
I mean to some extent this is it's a it's an extremely important time for Tesla because look there's obviously competition across the board. Some could argue Nvidia is a competitor but I just believe when it comes to the autonomous world they're going to own 80% of the market when it's all said and done them but 30 cities you need robo taxis this year it's integral you know in terms of the uh the Tesla story.
Yeah, I mean I know they started in Austin. That was one of the reasons why you were so positive on this name because they saw that and you said 30 cities coming for the robo taxi. You had two trillion market cap over the coming year. Bull case scenario, three trillion market cap by the end of 2026.
Let me ask you this. So the robo taxi story, is it rolling out as quickly as you had hoped? And will people really be comfortable going in a robo taxi? It's like asking, are you comfortable using stable coin in retail? Um, how quickly do you think that this could be adopted? I mean, I don't know that I'm ready to get into a car without a driver.
Are you? And how quickly these robo Yeah, I know you are. You probably already rode in it 20 times, but are people going to adopt this? And are is this roll out going as quickly as you would hope? And if not, would you get on the phone and tell and tell Musk, "Spe it up?" Look, but see, I view it as I like that they're doing it brick by brick carefully because you're dealing with lives, right?
So, they got to make sure that everything is right and and I'd rather think it's more important in terms of the way they build it out, making sure it's safe and making sure that as a geospace area expands, it's in the right way rather than just quick. They look awesome's expanded nicely. We're going to go to California, but like I said, like you really want to see 30 cities to get to scale on the robo taxi side.
Uh is why do people say Whimo's better than Tesla? And is that not true? Look, Whimo's had a huge it's had a huge start and they have a phenomenal technology. My whole thing is that Tesla no one will match the scale and scope of robo taxis eventually. But look, like I've said about Whimo, Whimo is another reason to be bullish in Alphabet.
It's not factored in. Tesla's fleet generates more driving data in a single weekend than Whimo collects in an entire quarter. And that advantage compounds every single day. Sedan's making his prediction about robo taxis in dozens of cities by 2026. And yeah, that sounds very aggressive, but the thing people keep overlooking is the data advantage Tesla has already built.
Every Tesla on the road is feeding the system. Rain, snow, weird intersections, someone running a red light at 2 a.m. The system has seen it all and learned from it. Whimo's fleet is so small relative to Tesla's that when analysts try to put them on the same chart, Whimo barely registers. The scale difference is hard to overstate, and this is why timelines are really important.
Tesla adds massive mileage every day just from people commuting to work. So when you hear Dan say robo taxis across dozens of cities, he's looking at a company that's already solving edge cases other companies haven't even encountered yet. Whimo is building a data set Tesla already has, and the gap keeps growing because Tesla's fleet keeps driving and growing at the same time.
The economics are pretty interesting, too. A dedicated robo taxi costs way less than a traditional car with a human driver's salary attached. If they can operate at a fraction of current ride sharing costs and still charge competitive prices, the margins get attractive quickly. When you've got that kind of math and you've got the factory capacity to scale production, all you really need is a regulatory green light and enough safety data to back it up.
The thing that makes Tesla's approach fundamentally different is it's a generalized solution. Dan touched on this when he said it tends to work everywhere. Tesla's vehicles have learned how to drive, period. They don't need lidar. They don't need pre-mapped cities. They don't need remote drivers watching every move.
And that's what makes the multi-ity prediction actually believable. Once it works in Austin, it works in Phoenix. Once it works in Phoenix, it works in Dallas. The rollout becomes a logistics problem, not an engineering problem. So Tesla's working towards deploying autonomous vehicles at scale. But there's another physical AI bet they're making that could end up being even larger over time.
Boston Dynamics showed up on 60 Minutes demonstrating their humanoid robot working in the Hyondai factory. And the CEO is talking about this moment like it's a major breakthrough. But when you actually look at what they're showing versus Tesla and what they're doing with Optimus, the competitive dynamic looks a lot like the robo taxi situation.
Best place to start with all of the abstract discussion on humanoids this week. Frankly, what is it that's changed that's allowed you now to say we have a production version of a humanoid robot? Well, we've been building robots a long time, and so we've figured a lot out about how to a build a reliable machine that you could massproduce, and that's what we've introduced at CES this week.
We're showing our research version. It's been running. Uh that's the 2-hour wait to get into our booth. We've been running demos all week long. Uh but we announced our our mass production version that we uh we're beginning to build out now. And uh we're going to scale production with Hyundai factories so that we'll be building these in the tens of thousands by 2028.
There is always going to be the question, how real is this? Right? So you and Hyundai are partners. They're a customer, but you're also working together to continually test. What is it that Alice is going to be doing in that facility? Is it revenue generating? In other words, is this a real commercial deployment?
Absolutely. So look, our humanoid is our third commercial deployment. We've already got two robots out generating revenue today. And so the commercial opportunity is real. What's special about Hyundai is that they're both a a producer. They can help us massproduce. They can also be a consumer uh to use robots profitably in their factories.
you know, they've defined their own softwaredefined factory uh initiative and robots are going to be a part of that initiative to make robot uh make their factories more flexible and productive. The brain of Atlas uh at least in in the inference context is is NVIDIA. Um but a lot of this work you've done yourself on the software side in particular.
Where would you say Boston Dynamics's core competency is and and how Atlas reflects that core competency? We've always been the uh the whole body coordinated motion first, right? We basically introduce robots like humanoids that could really in an agile and coordinated way move around in the world. And so our core competency is in sort of the the motion control and the low-level stuff uh that's in close proximity to the hardware.
We're going to be teaming with Google Deep Mind for the high level cognitive functionality, right? So that tasks that go over minutes uh or even longer will be orchestrated by Gemini Robotics. So we're going to kind of split the low-level motor control and uh and the high level cognitive control. How you train Atlas in the first instance particularly in the industrial use case is fascinating.
There is a great emphasis on virtual worlds, digital twins. I think the reality is, you know, tell operation plays a role here, but how complex is it to get the humanoid robot to interact in a factory facility, taking into account the actual physics of that's being applied to it in that environment.
Well, this is where AI has really provided a breakthrough. We can use simulation tools to uh coordinate the low-level motion of the robot. So, reinforcement learning. Uh we use teley operation. If you can teley operate a robot to do a task in a factory, you can pretty much get an autonomous policy doing that task.
So we've already demonstrated that multiple times. What we need to do now is generalize that so that in a day or two you can train a whole new task because ultimately humanoids are going to need to do hundreds of tasks in a factory. So there in lies the question like right now what is the problem we're solving for?
literally labor shortages which Jensen's been talking about all week or productivity efficiency gains you know what is it that you're selling with your products well the high level goal is to onshore manufacturing that requires enhancements in labor productivity and I think robotics have a have a strong role to play uh in that in that context you know uh demographics are uniform the world uh across we're seeing declining populations it's difficult for manufacturers to get the talent that they want.
So, the robotics are going to take the dull, dirty, dangerous parts of the jobs. The humans are still going to do the high level uh cognitive high value tasks. Boston Dynamics makes impressive video demos, but Tesla's building robots designed to be economically useful from day one. And that's a fundamentally different approach.
So, the Boston Dynamic clip looks cool, right? The robot can twist. It can turn in unusual ways. It can stand up from weird positions. But if you watch closely, what useful work is it actually doing? It's picking up something long and putting on a station. That's basically a vice grip with legs. And the robots they showed were completely coordinated in a low traffic corner of the factory.
That's not exactly integrated into the production line yet. Here's some history worth knowing. Google bought Boston Dynamics back in 2013 and a lot of people said that's it. Game over. They won robotics. Then Google sold them a few years later. Why? Well, because the leadership kept asking how they could make these robots economically feasible.
and Boston Dynamics didn't have a clear answer. The humanoid robot was more of a research project that made incredible YouTube videos than a product designed for mass deployment. Now, Hyundai owns them and they're showing similar demos on 60 Minutes. Meanwhile, Tesla's taking a different approach. Elon's being pretty open about this.
They're deliberately focused on making Optimus practical and affordable rather than maximizing physical capabilities. The insight is simple. You don't need a robot that does parkour. You need a robot that picks and places parts on an assembly line reliably for thousands of hours. Optimus is being designed with manual dexterity approaching a human hand with actuators built from scratch because nobody makes the components they need.
And the target production cost at scale is supposed to be comparable to a mid-range car and actually maybe even way closer to sort of $20,000. The parallel to robo taxis is worth exploring. Boston Dynamics has better brand recognition, more media coverage, way more impressive demo videos, but Tesla has the manufacturing capability, the AI training infrastructure, and a business model built around actually selling and deploying these things at volume.
They're building the robot assembly line now. They're planning to deploy units internally first, learn from every hour of operation, and iterate. Same playbook as FSD. Let the fleet teach the system. By the time competitors figure out how to scale production, Tesla wants to be several generations ahead and ramping volume.
Now, let's see Dan Iive's top five stock picks for 2026. When we think about the names overall, um Nvidia is your number one without a doubt. It seems to always be. Um Palunteer has been a name that you loved. Is that still in the top game for you and why? Yeah, it's one, Nicole, that's one of our top five names for 2026.
Like when I look at Palanteer just because in my view in terms on the software side that's going to continue to kind of play more and more of a role in terms of what Palunteer does because AIP this is really the start of just monetizing the enterprise and I think the AI story of Palunteer as you go more and more into the commercial look only 3% of companies have gone down the AI path that's why when I look at Palunteer we've talked about trillion dollar valuation next two to three years and they're playing a different game.
Now, I think others are going to continue to have success in software. MongoDB, Snowflake, a good example. And when you talked about um I know you mentioned cyber security too um but I wanted you mentioned the top five names 2026. Tell us right now so in case anybody's not sure, what are your top five names for 2026?
Yes, that's Tesla, Microsoft, Apple, Palunteer, and then in cyber security, Crowd Strike. .
Comments 0