Tesla (TSLA)’s car sales are stumbling, profits have nearly halved, yet its market cap stays above $1 trillion. The disconnect? A high-stakes bet on robotaxis and AI humanoid robots — not electric vehicles.
The EV slowdown is real. In 2025, U.S. EV sales fell 2%, but Tesla’s deliveries dropped 9% and profits were cut nearly in half. The Model S and Model X were discontinued. In Q1 2026, deliveries fell another 14% sequentially. The once-revolutionary carmaker is clearly being dragged down by its own EV baggage.
Still, some on Wall Street are wildly bullish. Ark Invest sees Tesla hitting $2,000 by the end of 2029 — based on a simple premise: Tesla is no longer a car company. It’s a robotics company.
Ark’s report is blunt: by 2029, robotaxis could account for ~90% of Tesla’s enterprise value, capturing a big slice of a projected $10 trillion global market.
This isn’t about today’s auto sales. It’s about manufacturing scale and data closed-loop capabilities. The Model Y has been the world’s best-selling car for three years. Tesla already produces over 5,000 vehicles a day. Ark notes that its Austin factory alone could out-produce the entire ride-hail fleet of urban Austin in just nine days. And the Cybercab — set to launch next year — targets 2–4 million units annually.
By contrast, traditional car sales would then contribute only a quarter of total revenue and ~10% of earnings, because robotaxi margins are far higher.
If robotaxis are Tesla’s first curve, Optimus humanoid robots are the second. Tesla recently halted Model S/X production to repurpose factory space for Optimus. Elon Musk says AI-powered humanoids will go on sale by late 2027.
Designed for tedious, repetitive tasks, Optimus could follow Tesla’s familiar playbook: sell hardware, then unlock high-margin recurring revenue via subscriptions and OTA updates. If Tesla builds a sticky ecosystem, it could lead the AI robotics market — projected to reach $375.8 billion by 2035, growing at 17% annually.
Three big questions remain.
First, EVs can’t be simply tossed aside. They are Tesla’s cash flow and the physical foundation for both robotaxis and robots. Without the mass-production know-how from the Model 3/Y, Tesla wouldn’t have the cost advantage for Cybercab or Optimus. Killing EVs would mean cutting off its own limbs.
Second, Musk’s timelines are notoriously unreliable. From FSD to the 4680 battery, his promises have often been years late. Ark’s $2,000 target depends heavily on that timeline. Any delay in Cybercab or Optimus would shake Tesla’s valuation.
Third, regulation looms large. Robotaxis face road rights, insurance, and liability hurdles in cities worldwide. Humanoid robots will confront safety and privacy rules at home and work.
Tesla isn’t really abandoning EVs — it’s pushing them to the back seat, turning them into the chassis and factory floor for its robotics ambitions. The future market cap will be built on robotaxis (a $10 trillion mobility market) and Optimus (hundreds of billions in AI robotics).
The real debate isn’t about the opportunity — it’s about the timeline. Ark may be too aggressive, but the direction looks right. As Ark itself said: “Tesla’s vertically integrated manufacturing offers a clear scaling advantage.” As long as that advantage holds, even shaky car sales won’t end Tesla’s long-term story.
