Is the robot fleet actually funded? Tesla's (TSLA) fiscal 2025 Form 10-K (period ended December 31, 2025), located through SEC filings, answers obliquely. The company says it is "working to develop and commercialize AI robots ('Bots') (including Optimus)," and that "We intend to leverage our current operations, in which we design, develop, manufacture, sell and lease high-performance fully electric vehicles and energy generation and storage systems... to achieve that objective." Read as a financing statement, that means Optimus rides on the existing balance sheet rather than a dedicated capital structure.

The numbers behind "current operations" are what make that a credible plan rather than a slogan. The FY2025 10-K reports total revenues of $94,827 million (down 3% from $97,690 million in 2024) and net cash provided by operating activities of $14,747 million for the year. The company ended 2025 with $44.06 billion in cash, cash equivalents and investments — $16.51 billion in cash and equivalents plus $27.55 billion in short-term investments — and an additional $6.43 billion of unused committed credit. That is the "existing machine" Optimus is being funded from: a business generating roughly $14.7 billion of operating cash a year sitting on a $44 billion liquidity stack.

Segment disclosure, not the keynote. Because Optimus is funded from inside Tesla's broader operations, its cost does not surface as a clean, isolatable line — it is folded into the company's overall R&D and capital spending. And both of those moved sharply. Research and development expense rose 41% to $6,411 million in 2025 from $4,540 million in 2024, climbing from 5% to 7% of revenues; the 10-K notes R&D consists primarily of personnel costs, and Tesla explicitly says it is "applying our artificial intelligence learnings from self-driving technology to Bots, such as Optimus." Capital expenditures, meanwhile, fell to $8.53 billion in 2025 from $11.34 billion in 2024. The point for an analyst is that you cannot separate humanoid investment from vehicle and AI investment in these consolidated numbers — Optimus is one undisclosed claimant on a rising R&D line and a multi-billion-dollar capex budget.

“While this development requires significant cash investments and management resources, there is no guarantee this business will be successful. We have yet to commercialize Bots and cannot predict how demand for Bots will develop, either from commercial or consumer applications.”— Tesla, Inc. 10-K (FY2025) source

R&D capitalized is R&D hidden, and Optimus is a good example of why the phrase matters. A program funded from existing operations and described in qualitative 10-K language — the filing calls Bots "in a nascent industry that has yet to develop commercially" and "an important part of our business going forward" — gives the reader no discrete spend figure to track. The capex tell is therefore indirect: watch total R&D and capital expenditure trends, and treat Optimus as one of several programs drawing on them.

The forward guidance reinforces how much capacity is being built around AI broadly. Tesla tells investors it "currently expect[s] our capital expenditures to be in excess of $20 billion in 2026, driven by our AI initiatives, including investments in compute infrastructure and data centers, the expansion and ramp of our manufacturing and R&D production lines and facilities." A jump from $8.53 billion in 2025 to a guided figure above $20 billion in 2026 is a step-change in capital intensity — and the filing names AI compute and data centers first among the drivers. Optimus, FSD (Supervised), and Robotaxi all draw on that same compute and manufacturing build-out, which is precisely why the dollars cannot be split apart from the outside.

The strategic logic the filing offers is reuse, and reuse is what makes the internal-funding model coherent rather than merely convenient. The 10-K says Tesla is "applying our artificial intelligence learnings from self-driving technology to Bots, such as Optimus, a general purpose, autonomous humanoid robot in development," and that it is "capitalizing on our strengths in real-world AI data to advance the development of Optimus." In capex terms, that is the argument that the compute infrastructure, the AI-training pipeline, and the real-world data already being paid for by the vehicle and FSD programs do double duty for the humanoid — so the marginal cost of Optimus is genuinely lower than a standalone robotics company's, and genuinely impossible to isolate. The same $20-billion-plus 2026 capex line that builds compute for Robotaxi and FSD is the line Optimus draws from, which is exactly why no clean Optimus number exists or, on this logic, should.

The honest limit: funding a program internally is a sign of confidence and scale, not a red flag. Tesla's MD&A states plainly that "Overall growth has allowed our business to generally fund itself, and we will continue to make critical high-value investments while maintaining a strong balance sheet." A company throwing off $14.7 billion of operating cash and holding $44 billion in liquidity can absorb a humanoid program without a separate raise. But the same internal funding that de-risks financing also removes the disclosure that would let an outsider size the bet — and the filing is explicit that there is "no guarantee this business will be successful."

The takeaway for the capex desk: Optimus is funded the Tesla way — from the existing machine, against a $44 billion liquidity base and a capex budget guided above $20 billion for 2026. The 10-K tells you that plainly while telling you almost nothing about the dollar amount going specifically to the humanoid. The financing question is answered; the sizing question is, by design, not.