The Gross-Margin Problem: Why AI Does Not Behave Like Traditional Software
AI revenue can carry significant model-serving and infrastructure costs. Gross margin reveals whether growth behaves like software or compute resale.
Gross margin exposes the difference between software and AI software
The first step is to define the metric precisely because finance terms that sound intuitive often have specific accounting boundaries. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. Microsoft reported that its cloud gross-margin percentage fell to 66% in fiscal 2026, with AI infrastructure investments and growing AI product usage among the pressures. [1]
A useful analytical habit is to separate operating metrics from accounting statements. Operating metrics can be excellent leading indicators, but they often omit financing structure, depreciation, stock compensation, tax, working capital, or the capital needed to sustain growth.
AI does not become free after the product is built
Labels are useful only after the underlying calculation is understood.
Microsoft’s cloud margin shows AI usage has a real cost
The current AI market provides unusually vivid evidence because companies are scaling revenue, compute, and capital commitments at the same time. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. CoreWeave describes power, rent, data-center operations, and depreciation as direct components of cost of revenue. [2]
The second habit is to ask what happens when usage doubles. If revenue doubles while direct serving cost rises almost as fast, scale may improve the headline without creating much operating leverage. If cost grows much more slowly, the same growth can produce powerful margin expansion.
Physical infrastructure can still produce software-like economics
A dramatic growth rate can coexist with weak unit economics.
CoreWeave places power and depreciation directly into cost of revenue
The mechanism matters: the same headline number can imply very different economics depending on what sits above or below it in the financial statements. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. Nvidia’s fiscal 2027 second-quarter gross margin was 75%, showing that upstream AI infrastructure can retain strong economics even while selling physical systems. [3]
A third distinction is timing. Accounting can spread some costs across years, recognize some revenue over contract periods, and exclude certain items from management-defined measures. Cash, however, moves when suppliers, employees, lenders, and infrastructure vendors are actually paid.
Training cost may sit outside a reported gross-margin figure
Cash and accrual accounting answer different timing questions.
Nvidia demonstrates that infrastructure can still carry high margins
AI intensifies the issue because model serving, infrastructure, research, and strategic financing introduce costs that ordinary software companies could often ignore. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. Anthropic reportedly exceeds 80% gross margin before accounting for partner revenue sharing and training costs, demonstrating why gross-margin definitions need careful reading. [4]
AI also makes capital structure part of product strategy. Companies with wealthy parents, strategic cloud partners, customer prepayments, or public-market access can finance expensive capacity years before a smaller competitor could. That can alter both market share and reported economics.
Efficiency improvements can compound across every customer request
AI scale magnifies small accounting assumptions into large valuation differences.
Frontier-lab gross margin can exclude economically important costs
Comparisons are useful only when the underlying definitions match. Two companies can use the same label while measuring different economic realities. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. AI software can therefore look like several businesses at once: high-margin subscription software, metered cloud infrastructure, and research-intensive model development. [5]
Definitions become especially important in private markets because investors often receive operating metrics without a full public filing. ARR, adjusted operating income, or gross margin may be informative, but an outsider may not see every exclusion or balance-sheet obligation.
Model routing and inference efficiency become financial controls
For investors, the important question is how the metric connects to future cash generation rather than whether the headline number looks large. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. Microsoft reported that its cloud gross-margin percentage fell to 66% in fiscal 2026, with AI infrastructure investments and growing AI product usage among the pressures. [1]
The best comparison therefore follows the money from customer payment to gross profit, operating expense, interest, tax, capital expenditure, and finally free cash flow. A metric is useful to the extent that it helps explain one part of that chain without pretending to be the whole chain.
Product mix determines whether AI margins expand or contract
Technology history repeatedly shows that growth metrics become less persuasive once markets mature and financing is no longer abundant. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. CoreWeave describes power, rent, data-center operations, and depreciation as direct components of cost of revenue. [2]
Valuation adds a future-tense layer. Markets can rationally pay for growth before current profit exists, but the price ultimately assumes that future revenue will convert into margins and cash after all required investment. The more capital intensive the model, the harder that conversion becomes.
Gross margin is the first real test of scalable AI economics
The durable interpretation is therefore the one that survives reconciliation to revenue, expense, cash flow, and capital requirements. Gross margin is where the economics of AI become visible because every inference, GPU hour, and cloud commitment can sit between revenue and gross profit. Nvidia’s fiscal 2027 second-quarter gross margin was 75%, showing that upstream AI infrastructure can retain strong economics even while selling physical systems. [3]
For CodeHistory, the larger historical point is that AI has not abolished finance. It has made old concepts—revenue quality, depreciation, operating leverage, dilution, cash flow, and cost of capital—more important because the sums involved are so much larger.
Works Cited
- 01Microsoft FY2026 10-K sec.gov
- 02CoreWeave FY2025 10-K sec.gov
- 03Nvidia Q2 FY2027 10-Q sec.gov
- 04Reuters — Anthropic Adjusted Profitability reuters.com
- 05
CodeHistory is a living archive. Citations document the evidence used for this edition; later evidence may refine the account.
Submit a research lead