Tesla is selling more vehicles again. Revenue is moving in the right direction. Its energy business is still growing.
Yet the company burned through cash.
That uncomfortable mix appeared in Tesla’s second-quarter 2026 results, where a sharp rise in capital spending pushed free cash flow into negative territory for the first time in more than two years.
The money is not disappearing into routine factory maintenance. Tesla is funding artificial intelligence infrastructure, robotaxi production, autonomous-driving development, battery capacity and humanoid robots. Elon Musk is trying to turn the electric vehicle maker into something much broader.
That transformation is becoming expensive.
Tesla Reports Negative Free Cash Flow as Investment Accelerates
Tesla recorded negative free cash flow of around $1.1 billion during the quarter. The company still ended the period with approximately $43.5 billion in cash and investments, so this is not an immediate liquidity crisis.
It is a warning light, though.
Free cash flow shows how much money remains after a company covers operating costs and capital investments. Tesla’s core businesses generated cash, but not enough to keep pace with the amount being poured into new facilities, computing systems and manufacturing programs.
Capital expenditure reached roughly $5.8 billion during the quarter, an increase of about 142% from the same period a year earlier. Tesla expects its full-year capital spending to exceed $25 billion, according to reports surrounding the company’s latest earnings release.
That is a lot of money for projects that may take years to produce meaningful returns.
AI Infrastructure Is Becoming One of Tesla’s Biggest Expenses
Tesla no longer talks about artificial intelligence as a side feature attached to its vehicles. AI now sits close to the centre of the company’s investment strategy.
The company is expanding the computing infrastructure used to train its autonomous-driving systems and robotics technology. That includes work connected to its Cortex computing facilities, data-centre capacity and increasingly powerful training clusters.
Tesla said its AI compute capacity in Texas had expanded to around 250 megawatts, with plans to raise that figure further. The infrastructure will support Full Self-Driving development, robotaxi services and Optimus humanoid robots.
None of this is cheap.
AI systems need advanced chips, electricity, cooling equipment, networking hardware and buildings large enough to house the machinery. Training models for real-world driving also requires enormous amounts of video data and repeated testing.
Tesla is effectively building the computing backbone for several future businesses at once.
The Robotaxi Push Is Moving From Promises to Manufacturing
Robotaxis have been part of Elon Musk’s Tesla story for years. The difference now is that the company is spending heavily to turn the idea into an operating business.
Tesla has begun preparing production for its purpose-built Cybercab at its Texas factory. It has also expanded limited robotaxi services into additional locations, including parts of Florida and Texas.
The rollout remains small compared with Musk’s previous predictions. Tesla once suggested autonomous vehicles could rapidly cover large portions of the United States. The real deployment has been slower, more controlled and closely watched by regulators.
Still, the company is moving beyond demonstrations.
Manufacturing Cybercabs, building fleet-management systems and operating a commercial ride service require more than better driving software. Tesla needs vehicles, charging infrastructure, maintenance networks, remote support and regulatory approval in every market it enters.
That explains part of the rising spending. It also raises the stakes.
Tesla’s Vehicle Business Recovered, but Margins Remain Tight
The cash burn arrived during what otherwise looked like a stronger quarter for Tesla’s automotive business.
The company delivered 480,126 vehicles, about 25% more than during the same quarter of 2025. Revenue increased 26% to approximately $28.2 billion, beating Wall Street expectations. Tesla also passed $100 billion in revenue over the previous 12 months for the first time.
Those numbers sound solid.
Profitability told a less comfortable story. Tesla reported net income of about $1.11 billion, while adjusted earnings fell below market expectations. Discounts and promotional pricing helped support vehicle demand, but they also placed pressure on margins.
Automotive gross margin, excluding regulatory credit revenue, stood at 16.3%. That was higher than the comparable quarter of 2025 but lower than the 19.2% reported during the first quarter of 2026.
Tesla can sell more cars and still feel squeezed when each vehicle produces less profit.
Optimus Is Taking Over Space Once Used for Premium Cars
Tesla’s spending shift is visible inside its factories too.
The company has ended production of the Model S and Model X and is repurposing part of its Fremont facility for Optimus humanoid robot manufacturing. Construction for the robot production system has already started.
That decision says plenty about where Musk thinks Tesla’s future lies.
The Model S helped establish Tesla as a serious electric vehicle company. The Model X became one of its most recognisable premium products. Replacing their production footprint with humanoid robots is not a minor adjustment to the product lineup.
Tesla is betting that Optimus could eventually become a larger business than its cars.
For now, however, the robots remain a cost centre. Tesla must develop the hardware, train the AI, establish manufacturing processes and persuade businesses that humanoid machines can perform useful work safely and consistently.
Tesla’s Energy Business Provides Some Breathing Room
Not everything outside the vehicle business is draining cash.
Tesla’s energy generation and storage division produced approximately $3.1 billion in quarterly revenue, up around 13% from the previous year. Battery storage deployments also returned to growth, giving the company another source of revenue as automotive margins fluctuate.
The energy division matters because Tesla’s AI plans need funding.
Vehicle sales still provide most of the company’s revenue, but energy storage could become a more dependable contributor as electricity demand rises around data centres, renewable power projects and national grids.
There is an interesting loop here. AI data centres consume huge amounts of electricity, while Tesla sells battery systems designed to help manage power supply.
Still, energy growth alone may not cover the scale of Tesla’s current investment program.
Investors Now Need Proof That Tesla’s AI Bet Can Pay
Tesla has enough cash to keep spending. The harder question is whether the company can turn that spending into profitable businesses before investment demands climb even further.
Robotaxis could create recurring transportation revenue. Full Self-Driving subscriptions could improve software margins. Optimus robots might open an entirely new market. Tesla’s custom AI infrastructure could also reduce its reliance on outside technology suppliers.
Could is doing plenty of work in those sentences.
Tesla’s existing vehicle operation remains exposed to pricing pressure, intense competition and changes in government incentives. Regulatory credit revenue is becoming less dependable, while autonomous-driving projects continue to face safety and approval hurdles.
The company is entering its most capital-intensive period at exactly the moment investors want clearer evidence that AI, robotaxis and robotics can produce real commercial returns.
Tesla Is Starting to Look Less Like a Traditional Car Company
For years, Musk has argued that viewing Tesla as an automaker misses the point.
The latest spending numbers make that argument easier to understand. Tesla is building cars, charging networks, energy storage systems, AI supercomputers, autonomous taxis, semiconductor projects and humanoid robots.
It is a strange collection of businesses. Maybe that is the opportunity. Maybe it is the problem.
The second-quarter cash burn does not prove Tesla’s strategy is failing. Large technology transitions often require years of heavy investment before profits appear.
It does show that the AI transformation is no longer just a presentation slide or a distant promise.
Tesla is paying for it now.
Sources
- Reuters: Elon Musk’s Tesla posts cash burn as capex surges on AI, robotaxi push
- Tesla Q2 2026 Update: Tesla Q2 2026 Update
- Tesla Investor Relations: Tesla Releases Second Quarter 2026 Financial Results
- The Verge: Tesla Q2 2026 earnings, revenue, profit and sales