Before beginning this paper, I want to make one point clear. After my weekend video, it became obvious to me that many investors hold deeply entrenched views about Tesla and Elon Musk. As you read, I ask that you set those aside. This is not about personalities, politics or stock debates, it is an AI paper. The focus here is on what is truly required, from a brain perspective, to achieve embodied AI and why Tesla must be understood in that context today.
Everyone loves the glossy clips of humanoid robots or autonomous cars gliding through city streets but they are illusions of progress at this point. What we’re seeing are specialized robots in humanoid costumes, machines that don’t yet see, they memorize. Like Waymo’s hyper-mapped, choreographed approach, they perform dog tricks in controlled settings, not genuine intelligence. True humanoids and Tesla’s vision-first bet on autonomy require photons and perception, turning raw light into understanding so the AI can adapt to unstructured, unpredictable environments, like Mars. Until machines can learn this way, everything else is choreography, not revolution.
This is why Tesla’s work on car vision is so important. The importance of this is why everyone needs to check your cognitive bias at the door as an investor. Every mile driven with cameras trains AI to interpret the world through photons, the same raw material humanoids will need to navigate factories, homes, and cities. The perception breakthroughs that make a car understand depth, distance, and intent directly transfer to embodied AI. In this sense, solving vision for cars is solving vision for humanoids, and Tesla’s data flywheel gives it a head start on building the first machines that don’t just repeat tricks, but actually see and learn. Now, back to the regularly scheduled program.
Executive Summary
A year ago, investor Gavin Baker warned that Tesla’s Full Self-Driving program would soon force skeptics to eat their words, predicting that autonomy would flip Tesla’s economics in a way “abjectly humiliating” to doubters. At the time, attention was fixated on ChatGPT, NVIDIA, and large language models. But Baker’s point that the true trillion-dollar markets lie in autonomy and physical AI has only grown sharper with time.
Today, Tesla’s robotaxi rollout is no longer a futuristic dream. Elon Musk has promised limited service in Austin this year and “millions of Teslas operating fully autonomously” by 2026. We all know he has a history of overpromising and underdelivering, so don’t get focused on the exact dates. Unlike competitors such as Waymo, Tesla’s vision-first, global approach relies on billions of real-world driving miles, creating a data flywheel that scales like GPT models. Each mile makes the system smarter, pushing autonomy from theory to deployment. This vision-first foundation is not just a tactical choice—it is the only path that bridges self-driving cars to true humanoids.
The economics are staggering. Robotaxis transform Teslas from depreciating assets into income-generating machines, redefining the auto industry’s margin profile. A car that once sat idle can now operate as a revenue engine, seeding Tesla’s network at virtually no capital cost. Combined with Tesla’s $16.5 billion Samsung chip deal to secure compute capacity, the company is building an ecosystem where AI, data, and manufacturing reinforce one another in a self-reinforcing loop.
This is more than just a mobility story. As NVIDIA’s Jensen Huang framed it, the next AI frontier is embodiment, intelligence moving from screens into the physical world. Robotaxis are the first scaled proof of embodied AI, robots on wheels mastering physics in real time. Their success paves the way for humanoid robotics, logistics automation, and an AI-driven industrial revolution that could rival the scale of the internet or the iPhone.
Tesla’s robotaxi network represents the hinge point where AI stops being abstract software and becomes the operating system of the physical economy. Just as electricity unlocked Edison’s light bulb and Ford’s assembly line powered a century of industrial growth, robotaxis may be the catalyst that turns embodied AI from speculation into reality. For investors and policymakers alike, the message is clear: the AI cycle has only just begun, and Tesla sits at the vanguard of its physical expression.
Roadmap
Normally, I would not include a roadmap but this is a long paper due to the importance I see for the potential secular regime shift I see starting now and how Robotaxis would be the ChatGPT moment of the rise of intelligent machines if successful.
From FSD Skepticism to the Economics of Autonomy
Skepticism has long surrounded Tesla’s Full Self-Driving program, often dismissed as hype. Yet Gavin Baker’s early insights highlighted how autonomy could transform Tesla’s economics, reframing it from a car company to an AI company.
The Embodiment Phase: From LLMs to Physical AI
The conversation around AI has shifted from text and language models to the physical world. Jensen Huang and others frame this “embodiment phase” as a multitrillion-dollar opportunity where robots, vehicles, and machines become the next great platforms.
Why Robotaxis Are Tesla’s “Light Bulb Moment”
Just as Edison’s light bulb only mattered once electricity scaled, robotaxis may prove to be Tesla’s catalytic moment. They are not simply a new product—they are the first scaled proof that embodied AI can function in the messy realities of the physical world.
Vision vs. Maps: Tesla’s Data Flywheel Advantage
The philosophical divide between Tesla and Waymo reveals the future of autonomy. While Waymo depends on geofenced maps and expensive sensors, Tesla’s vision-first approach turns billions of real-world driving miles into the most valuable dataset for embodied AI.
Beyond Cars: Robotaxis as the Foundation of Embodied AI
Robotaxis are not an endpoint, but a beginning. They are the scaffolding for humanoids, logistics systems, and embodied AI agents that will redefine industries and reshape the global economy.
From FSD Skepticism to the Economics of Autonomy
I remember listening to an Invest Like the Best episode in August a year ago with investor Gavin Baker. I was walking past the United Nations building, surrounded by diplomats and flags waving in the wind. At the time, the conversation was dominated by large language models and NVIDIA’s role in powering the AI boom. Robotics and autonomy felt like distant topics, mentioned almost as an afterthought compared to the excitement around ChatGPT and the scaling of compute. Yet buried toward the end of the discussion was a memorable set of remarks about Tesla’s Full Self-Driving program, comments that didn’t grab the spotlight then, but I remembered because of the importance.
He said:
“If autonomy works; if Tesla even partially solves it, the company’s economics change fundamentally. The margin profile could flip overnight because everything about software leverage and data scale we see in LLMs will apply to cars, but most people are still thinking of Tesla as an auto manufacturer, not as an AI company.”
However, it was this line that I will never forget
“This is going to be a reality in a way that it’s abjectly humiliating to everyone who is an FSD skeptic in the next 12 to 18 months, maybe in the next six months. And I have never been willing to make a prediction like that before.”
Skepticism around FSD, or AI more broadly, is hardly surprising. Traditional auto analysts are trained to model units, margins, and supply chains, just as economists and macro investors are trained to measure the physical world. But Tesla’s robotaxi vision forces both into unfamiliar terrain. AI is not simply another technology to slot into a model; it is thinking embedded in a machine, a shift from mechanics to cognition. Valuing a company at the intersection of cars and artificial intelligence is therefore a conundrum, one that challenges not only analysts but every investor confronting how to price the future of embodied AI.
The Embodiment Phase: From LLMs to Physical AI
We are now 12 months from that interview in August 2024. Baker noted that while enthusiasm at the time was all about ChatGPT and LLMs, the truly enormous market caps may come from the companies that successfully execute on autonomy and physical AI.
Then in January, NVIDIA CEO Jensen Huang sharpened that perspective. At GTC and CES 2025, he described the next great leap for artificial intelligence as the “embodiment phase” when intelligence moves from the screen into the physical world. He called this shift “the decade of autonomous vehicles, robots, and autonomous machines,” framing it as a multitrillion-dollar opportunity.
For Huang, future AI systems must master the laws of physics, not just patterns in data: “The next wave… requires us to understand things like the laws of physics, friction, inertia, and cause and effect.” In his view, embodiment is not only a technological inflection point but also a massive economic one transforming industries like logistics, transportation, and manufacturing, which together account for tens of trillions of dollars in global output. Most importantly, he stated that the ChatGPT moment for “Physical AI” was just around the corner.
It is very surreal to be doing research all week on AI impacting the physical world in the same week I received a wave of messages about the latest media claims that AI is a bubble. It’s striking how predictably this cycle plays out: portfolios dip, investors search for a narrative, and the press obliges with articles framing AI as overheated and nearing an end. In reality, we remain in the early stages of AI’s progress , in adoption, investment opportunities, and ultimately disruption. As I argued last week in my paper, The Academic Fed vs. the Inflation Target of the Future, we are not witnessing the end of an AI cycle but rather the beginning of a broader regime shift in which AI-driven investments will play a central role in fiscal and monetary policy decisions for the years to come.
Why Robotaxis Are Tesla’s “Light Bulb Moment”
At the same time these calls came in, I was deep in research on what I believe will be the clearest expression of this shift if it works: Tesla’s Robotaxi strategy. Tesla has been the worst-performing member of the “Mag 7” this year and Elon Musk remains one of the most polarizing figures in the world which makes this so interesting. Over the past few weeks there have been multiple references to robotaxis in my daily research. It has been about the excitement building on what is happening in Austin and the increase in X posts from Elon related to it. However, it was when John Roque mentioned that he liked the chart of Tesla for a possible base breakout that forced me to expand my thoughts. When a chart can match up with a fundamental catalyst, I get much more interested especially when I think it can be a symbol for investors in an investment regime shift and that’s what led to this paper. In this case, for the chart to look good when the news for Tesla and Elon has been so bad, even more reason to pay attention.
When you combine the chart with the steady drumbeat of podcasts, research, and posts about robotaxis, it reinforced my sense that the timing to focus on Tesla was now. For disclosure, I bought my first Tesla in 2012 after driving a friend’s and have only owned Teslas since. Yet despite being a loyal customer for over a decade, I’ve never really understood the stock or how to value it. Part of that comes from the nonstop Musk headlines, and part from the sheer improbability of one person running multiple large-scale businesses simultaneously. That tension, between my conviction as a customer and the market’s confusion, is what makes the robotaxi discussion so important right now.
I share this because the purpose of this paper is not to debate the stock, but to explain why I believe Tesla and robotaxis in particular may mark the official starting point of the robotics investment cycle: when embodied AI displaces the long-standing focus on software and infrastructure. Given embodied AI takes the ability to scale from a manufacturing perspective to drive excitement, Tesla seems like the only place it can come suddenly. The fact that Tesla’s robotaxi vision keeps surfacing on podcasts, in research, and across X suggests it’s time to pay attention. More importantly, these discussions, including Musk’s own posts, underscore that convergence is drawing near: from breakthroughs in FSD software and Austin’s successful pilot rollout to regulatory momentum in Texas, the system’s functionality in multiple countries, and techno-enthusiasts forecasting major expansion by next year. All of this validates the regime shift argument I outlined in my paper, From Cloud LLMs to Embodied AI: The Next Hardware Investment Frontier and Macro Regime Shift, where I posit that robotaxis will be the trigger for embodied AI’s emergence. To me, this isn’t Tesla’s moment, it’s hardware’s moment.
Vision vs. Maps: Tesla’s Data Flywheel Advantage
This view dovetails with the strategy Elon Musk has pursued with Tesla’s Full Self-Driving program and robotaxis. Where Waymo relies on a geo-fenced, LiDAR-heavy model optimized for structured grids and fair weather, Tesla adopted a video-first, neural-network, massive-compute approach. To simplify: Waymo drives like a student who can only handle the same well-mapped neighborhood, using expensive sensors and pre-loaded directions. Tesla drives like a teenager who learns by experience, using cameras and practice to get better anywhere in the world—even on roads it has never seen before.
Huang himself has acknowledged Tesla’s lead: “Tesla is far ahead in self-driving cars. But every single car, someday, will have to have autonomous capability. It’s safer. It’s more convenient. It’s more fun to drive.” Musk’s bet was that real-world video data — collected from millions of Teslas in the wild — would scale better than costly mapping and sensor systems. As Baker and others have noted, Tesla’s FSD scaling resembles GPT’s trajectory: starting with rudimentary outputs but improving exponentially as more data and compute are added.
Huang’s framing of embodiment makes clear why Musk’s approach may prove decisive. Robotaxis aren’t just cars, they’re the first wave of embodied AI agents, robots on wheels trained by exposure to the chaotic realities of the physical world. Waymo’s strategy may deliver safe performance in constrained environments, but Tesla’s global fleet and video-first neural networks embody exactly the type of physics-grounded intelligence Huang argues is necessary for the next industrial revolution. In Huang’s words, “autonomous vehicles will become the first multitrillion-dollar robotics industry.”
Musk’s robotaxis, if successful, are not the endpoint, they are the gateway to humanoids and general-purpose robotics, the first proof that embodied intelligence can scale, commercialize, and transform the economy. Without vision, these machines do not have thinking brains; they only have memory.
Gavin Baker pointed out that every Tesla equipped with AI hardware is, in essence, a robot on wheels collecting data, learning, and progressing along scaling laws similar to GPT. He predicted that skeptics would soon be “abjectly humiliated” as Tesla’s FSD accelerated from a GPT-2 level of capability to GPT-4 quality within a short timeframe, thanks to its unmatched dataset and compute buildout. A year later, within the timeline of his prediction a year ago, as Tesla prepares the rollout of its robotaxi network, those offhand comments stand as a reminder of how quickly narratives can shift. What once seemed like a side note is now at the center of one of the most important technological and economic stories unfolding today.
That connection back to Gavin Baker’s interview resurfaced last week while listening to a recent episode of Moonshots with Peter Diamandis, where they discussed Tesla’s Austin rollout, something that would’ve been unthinkable just a year ago. What struck me most is how Tesla’s robotaxi has shifted from science fiction to something that feels imminently close.
Musk talked about this during Tesla’s Q1 2025 earnings call on April 22. He announced that Tesla would launch a limited robotaxi service in Austin, Texas starting in June 2025, using a small fleet of Model Ys “maybe 10 to 20 vehicles on day one” with plans to scale rapidly. He projected expansion to other cities by the end of 2025 and “millions of Teslas operating fully autonomously in the second half of next year” (2026). He emphasized Tesla’s camera-only approach, rejecting lidar and radar in favor of scalable vision systems, acknowledging that conditions differ by region but betting on a universal learning system.
The Moonshots panel didn’t dismiss Musk’s forecast as wishful thinking; they emphasized that the real bottleneck is no longer technology, but regulation.
That’s why Texas’s decision to issue Tesla a statewide rideshare license is so pivotal. The Texas Department of Licensing and Regulation has granted Tesla Robotaxi LLC a Transportation Network Company (TNC) permit, valid through August 2026, enabling supervised and fully driverless robotaxis across the state . The hardware is ready, AI models are improving weekly, and these cars are learning in real-time on public roads. The brain may not be ready yet but it will be learning at a faster rate for the next year.
The real enabler here is Tesla’s data flywheel. Unlike competitors such as Waymo that rely on painstaking mapping and specialized sensors, Tesla collects billions of miles of real-world driving data from its fleet every day. Each near-miss, every odd traffic scenario, every edge case gets fed back into the system, making the next update smarter and safer. It’s this fleet-scale self-learning capability, powered by neural nets, that makes Musk’s timeline plausible. With a major FSD update expected soon, 10x larger and more precise, Tesla’s software is poised for another leap forward.
But the most exciting piece isn’t just the technology; it’s the economics. Musk envisions a world where a Tesla isn’t just a car, but a productive asset. Owners will be able to rent out their cars when they’re not using them, turning idle hours into income. For the auto industry, which has long depended on selling depreciating assets, this flips the model on its head. Cars become revenue-generating machines, injecting innovation and capital into a sector that has stagnated for decades.
Behind the scenes, Tesla is also sending a expectation signal by recently securing the compute power needed to make this all real. A $16.5 billion deal with Samsung locks in production of Tesla’s Dojo AI chips, ensuring enough supply to train and run the massive models powering autonomy. This is about more than cars: it strengthens Tesla’s position in AI research, reduces reliance on Nvidia, and lays the groundwork for humanoid robotics. Robotaxis are just the tip of a much larger AI iceberg that Musk is methodically piecing together.
The importance of robotaxis comes into sharper focus when viewed through the broader lens of AI’s transition from bits to atoms. As the Morgan Stanley documentary with Adam Jonas AI Is About to Get Physical argues, artificial intelligence has already consumed much of the “knowledge economy,” disrupting occupations that revolve around text, data, and digital workflows. But the true frontier is the physical economy, transportation, logistics, manufacturing, sectors that collectively represent tens of trillions of dollars of global GDP. Robotaxis sit at this exact intersection: they are among the first large-scale deployments of AI systems that don’t just process symbols on a screen but act in and upon the physical world, navigating the laws of physics in real time.
This transition has echoes in history. Just as Edison’s light bulb became transformative only once electricity was scaled cheaply, autonomy will only reshape mobility once it is deployed at massive scale. If successful, the robotaxi is Tesla’s “light bulb moment.” At first glance, it looks like a marginal improvement over ride-hailing, a cheaper, driverless Uber. But the true significance lies in what it enables: vast amounts of real-world vision data feeding back into AI systems, closing the sim-to-real gap, and compounding improvements faster than any competitor. As the documentary puts it, “If you solve for autonomy for cars, you solve autonomy for everything.” Cars are not the endgame; they are the training ground for embodied AI.
Finally, the stakes extend beyond economics to national competitiveness. Carl Sagan’s warning about America hollowing out its manufacturing base was a reminder that nations ignore physical production at their peril. Embodied AI, of which robotaxis are the first commercial wave, represents both a new utility and a new battleground. Whoever controls the networks of autonomous machines will control not just mobility, but logistics, manufacturing, and even defense. In this sense, Tesla’s robotaxi network is more than a business model innovation. It is the opening move in a much larger contest over how intelligence will be embedded in the physical fabric of the global economy.
When asked about Tesla’s true moat, Musk has consistently framed it not as a single breakthrough but as the convergence of six interlocking verticals, Data, Robotics, Energy, AI, Manufacturing, and SpaceX, which together spell DREAMS. This isn’t accidental; it reflects how Musk has thought several steps ahead, building capabilities across the entire embodied-AI stack. Data fuels AI, AI powers robotics, robotics require energy, energy is scaled by advanced manufacturing, and SpaceX provides the connective tissue through resilient global connectivity. By designing this ecosystem in advance, Musk has ensured Tesla isn’t dependent on any one supplier, technology, or business line, instead, each vertical reinforces and amplifies the others.
Musk’s forward planning and literalized in that recent $16.5 billion multi-year agreement with Samsung to produce Tesla’s next-generation AI6 chips at a dedicated Texas fabrication plant, entirely focused on powering Tesla’s robotaxi fleet, Optimus humanoids, FSD systems, and Dojo data centers. Samsung’s facility, co-located near Tesla’s operations, enables Musk to play a direct role in maximizing manufacturing efficiency, even joking that he’ll personally “walk the line” to accelerate progress This deal is not incidental; it fortifies the Manufacturing and AI pillars of DREAMS, ensuring control over critical semiconductor supply and reinforcing Tesla’s vertical integration.
By embedding chip production into his ecosystem Musk has effectively locked in the hardware backbone that will drive both robotaxis and future embodied AI products. The AI6 deal exemplifies how Tesla is not merely planning for tomorrow, it is factory-building, chip-designing, and data-streaming its way into the future. That is why its position in the robotaxi race, and the broader world of embodied AI, is not just powerful, it is strategically a difficult moat to break for others if he gets ahead.
Beyond Cars: Robotaxis as the Foundation of Embodied AI
Taken together, the robotaxi rollout isn’t just about cheaper rides or cooler tech. It’s about reshaping the economics of mobility, creating a new class of productive assets, and positioning Tesla at the center of an AI-driven economy. Cars that drive themselves are the first act, but the data they generate, the chips that power them, and the AI that directs them all feed into Musk’s larger vision of autonomy extending from roads to factories to homes. In that sense, Tesla’s robotaxi isn’t simply a new business, it’s the scaffolding for an entirely new industrial revolution.
The economic implications of robotaxis are hard to overstate. By removing the human driver, Tesla unlocks a margin profile that rivals software companies rather than traditional automakers. At scale, a robotaxi could operate 16 hours a day, generating over $15,000 in monthly revenue. This transforms the car from a depreciating consumer good into an appreciating, income-generating asset. It also redefines Tesla’s competitive set: not other carmakers, but platform companies like Uber, Airbnb, or even Amazon, which leveraged network effects to dominate entire industries.
Tesla’s owner-network model magnifies this disruption. Instead of building and financing a centralized fleet like Waymo, Tesla would effectively crowdsource its robotaxi fleet from existing customers, who have every incentive to put their cars to work. For owners, a Tesla that earns $10,000 to $30,000 annually in robotaxi revenue becomes a financial asset as much as a personal vehicle. For Tesla, each additional sale seeds the network at no extra capital cost, creating a powerful adoption flywheel. This “rental-as-a-service” approach may prove as disruptive to transportation as Airbnb was to hospitality.
The contrast with Waymo is clear: its geofenced, sensor-heavy model may be safer today, but Tesla’s vision-first approach is built to scale globally.
Of course, this transformation will not come without societal consequences. Robotaxis directly challenge the livelihoods of millions of drivers who rely on ride-hailing or taxi services worldwide. Displacement at this scale will trigger political debates around labor, regulation, and economic equity. Regulators will also face thorny questions: who is liable in a crash, the owner or Tesla? Should autonomous fleets be capped to prevent congestion? Consumer trust will matter just as much. Edge cases like a robo-car caught in a police blockade or unable to navigate a construction site risk undermining adoption if they are perceived as common rather than rare.
What makes robotaxis more than just a transportation story is their role in the broader evolution of artificial intelligence. They are the first scaled proof that AI can safely and profitably operate in the physical world. Just as LLMs commoditized intelligence in language, robotaxis demonstrate how AI can commoditize mobility. The leap from “robots on wheels” to humanoids in warehouses, factories, and homes is not as far as it seems. Tesla’s robotaxi rollout, then, isn’t only about cheaper rides or higher margins it is the scaffolding for embodied AI, a foundation for the next industrial revolution where intelligence becomes physical, distributed, and economically indispensable.
Tesla’s robotaxi rollout makes clear that the center of gravity in AI is shifting. As Jensen Huang has emphasized, the next great leap for intelligence is embodiment — mastering the laws of physics, friction, inertia, and cause and effect. The AI Is About to Get Physical documentary sharpened the same point: AI has already consumed the “bits and bytes” of the knowledge economy, but the real frontier lies in the physical economy of atoms and photons. Robotaxis are the first large-scale proof that this transition is happening, a fleet of embodied agents learning in real time, operating within the constraints of physics, and generating the vision data that fuels exponential improvement. Just as large language models scaled through text, Tesla’s fleet scales through photons, feeding the world’s most valuable dataset for embodied AI. This is why robotaxis are not an endpoint but a beginning the trigger that shifts AI from the cloud to the physical world, and the scaffolding for the next industrial revolution.
Even if you still have doubts about the likelihood of success for Tesla, I think it is important to have some potential upside if it does move forward. In a recent research report from Cathie Wood and ARK Invest, it reinforces this point with scale and economics. In Austin, Tesla’s limited rollout showed how just ~200,000 vehicles could meet the city’s entire urban vehicle miles traveled (VMT) a fleet size far smaller than most expect, and one Tesla could manufacture in days. ARK projects that Tesla’s robotaxi business could represent ~90% of its enterprise value by 2029, capturing a meaningful share of what they estimate to be a $10 trillion global robotaxi opportunity. Adam Jonas talked about a similar opportunity is his video. Tesla’s vertically integrated manufacturing, data advantage (collecting 40x more daily real-world miles than Waymo), and lower cost per mile position it to expand rapidly across the U.S. and globally. Where Waymo scales slowly through geofenced markets and partnerships, Tesla can “flip the switch” across its owner network, turning millions of cars into productive assets overnight. This ability to manufacture at scale and instantly leverage that scale through data and AI reflects precisely what the AI Is About to Get Physical documentary described as Musk’s true moat: the convergence of D.R.E.A.M.S.: Data, Robotics, Energy, AI, Manufacturing, and SpaceX. Among these, manufacturing sits at the center, enabling Tesla to build the probes that collect the data, improve the AI, and then manufacture even more probes in a self-reinforcing loop.
This is the essence of the regime shift: from software to hardware, from bits to atoms, from simulated language to embodied intelligence. Tesla’s robotaxis represent the first commercial wave of embodied AI — proof that autonomy can scale, commercialize, and transform industries. The downstream effects will ripple far beyond mobility: into logistics, defense, urban planning, and humanoid robotics. The investment community has been fixated on cloud compute, LLMs, and infrastructure AI. But the real inflection point is happening on the streets of Austin and San Francisco, where robotaxis are no longer theory but reality. This is not just Tesla’s moment — it is hardware’s moment, and it signals the beginning of an industrial revolution driven by embodied AI.
The Moonshots with Peter Diamandis podcast put it best as to why now: “The reason we have self-driving now and not five years ago, 10 years ago, 20 years ago is purely because the AI neural net is self-evolving, self-learning. The mechanical parts have been there for plenty of years. It’s the intelligence that makes it actually work.”
What makes that intelligence possible today is the massive buildout of compute and data centers. Without the hyperscaler investment and the colossus of infrastructure behind AI, there would be no breakthrough in autonomy. This isn’t evidence of a bubble—it’s evidence of progress. We’re in the midst of the next stage, where scaling compute fuels scaling intelligence, and scaling intelligence unlocks entirely new industries like RoboTaxis which leads to humanoids and all forms of embodied AI. As Leonardo da Vinci once said, “Every step you take reveals a new horizon.”
Gavin Baker’s warning to skeptics should not be ignored, especially given Musk’s recent track record in AI. Elon Musk has an uncanny knack for scaling late-entry projects at exponential speed, xAI, founded in July 2023, is a prime example. In less than two years, xAI went from inception to the release of Grok 4 on July 9, 2025, catching up to (and in some benchmarks surpassing) far better-funded, longer-established AI labs. OpenAI, as an example which Musk helped start began in 2015. Recently released, Grok 4 harnessed xAI’s Colossus supercluster, now powered by 200,000 GPUs, and leveraged infrastructure and algorithmic innovations that boosted training efficiency sixfold to “refine Grok’s reasoning abilities at pretraining scale.” That compute backbone didn’t materialize slowly: Jensen Huang praised Musk’s team for building the Colossus cluster featuring 100,000 Nvidia H100 GPUs in just 19 days, a feat he called “superhuman,” given that similar projects typically take years. Musk’s playbook is clear: enter late, then scale overwhelmingly in both compute and capabilities accelerating progress in quarters, not years.
Just as the light bulb only mattered once electricity scaled and the assembly line reshaped the last industrial revolution, robotaxis may prove to be the catalyst for embodied AI, the moment when intelligence escapes the screen and transforms the physical world. For cognitively unbiased investors, this is Tesla’s true optionality: not in incremental car sales, but in capturing a share of a multi-trillion-dollar mobility market of the future that could rival the impact of AWS or the iPhone. Adoption won’t be linear, but once the flywheel of data, AI, and manufacturing turns, it will accelerate far faster than expected. This is not just the next chapter of Tesla’s story, it is the opening act of the embodied AI economy.
This is not about Musk, nor about debating Tesla as just another car company. It is about whether vision-first autonomy marks the moment when AI leaves the screen and enters the physical world. Robotaxis are not an endpoint, they are the first proof that embodied intelligence can scale, commercialize, and transform the economy. Investors who focus only on quarterly delivery numbers or Musk’s timelines risk missing the bigger truth: solving vision for cars is solving vision for humanoids, and that is the foundation of the embodied AI economy now coming into view.
