PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteBefore investing in an AI company, test whether it can repeatedly solve a valuable customer problem, earn revenue on terms that cover the cost of serving it, and fund its operations without relying indefinitely on new capital. Start with the company’s own latest filings and financial statements. An AI label, a growing sales figure, or a technically impressive model is not evidence by itself that the business is durable or that its shares are attractively valued.
1. Identify what the company sells—and who pays
Describe the business in one sentence: “The company sells [product or service] to [buyer] to solve [problem], and charges by [pricing method].” Use the company’s filings and product descriptions to fill in each part. If you cannot tell what the paying customer receives or which budget pays for it, treat that as an unresolved diligence question.
AI businesses can occupy different parts of the value chain. A company may sell model access, an end-user application, cloud or compute infrastructure, consulting and implementation, or a bundle of these. The distinction matters because buyers, costs, sales cycles, and sources of competitive advantage can differ. For a company with a broader software portfolio, separate revenue actually attributable to AI products from revenue merely associated with the company’s AI branding.
| Offer type | What to establish | Business-model question |
|---|---|---|
| Model or model access | Whether customers pay for API calls, licenses, hosting, or another form of access | Can the provider charge enough per unit of use to cover compute and other delivery costs? |
| Application | Which workflow the product changes and which team buys it | Does it become valuable and embedded in routine work, or is it easily replaced by a bundled alternative? |
| Infrastructure | What compute, hosting, data, or development capability is sold | Can usage and pricing grow without costs or capital needs growing just as quickly? |
| Consulting or implementation | How much project work is required before a customer reaches production | Can delivery be repeated efficiently, or does growth depend mainly on adding people to each engagement? |
| Bundled offering | Which components are included, separately priced, or required for deployment | Can the company identify which parts drive adoption, revenue, and margin? |
2. Reconstruct how revenue is earned
Do not rely on labels such as “subscription” or “annual recurring revenue” without checking the accounting and contract details. In the latest annual and quarterly filings, review the business description, management’s discussion and analysis, audited income and cash-flow statements, revenue-recognition footnotes, customer-concentration disclosures, and risk factors. Check contract length, renewal terms, how and when revenue is recognized, and whether hosting, support, or implementation is bundled.
#1 Best Overall
Classify reported revenue as subscriptions, usage-based consumption, licenses, services, or a mix. Then distinguish committed payments from revenue that depends on future usage, contract renewals, or project work. Backlog, bookings, remaining performance obligations, announcements, and recognized revenue are different measures; none should be treated automatically as interchangeable proof of customer demand.
C3.ai’s fiscal 2026 Form 10-K illustrates why a subscription label is not enough: it describes ratable and consumption-based subscription recognition, runtime fees, customer-hosted and vendor-hosted options, and cloud-provider hosting costs. The issuer reported that subscriptions accounted for 91% of total revenue in fiscal 2026, 84% in fiscal 2025, and 90% in fiscal 2024. Those percentages describe C3.ai’s reported revenue mix in those specific years, not a sector benchmark. Read C3.ai’s fiscal 2026 Form 10-K.
3. Test whether customers get lasting value
Look for evidence that customers move beyond trials and use the product in production. Where companies disclose it, examine renewals, repeat usage, expansion within existing accounts, deployment numbers, and customer outcomes. Ask whether the product is tied to a measurable business result—such as reducing a defined cost or improving a specific process—and whether there is evidence that the result persists after initial implementation.
- Separate production deployments from pilots and demonstrations.
- Compare renewals and expansion with new-customer wins; each indicates something different about customer retention and acquisition.
- Check whether implementation requires substantial customization, ongoing human review, or services from the vendor.
- Track customer concentration across reporting periods and read the associated risk disclosures.
- Ask what makes the product costly or difficult for a customer to replace, rather than assuming that AI use alone creates switching costs.
A few large contracts can support growth while also leaving results exposed if one customer reduces spending or does not renew. C3.ai identifies concentration and renewals as risks in its fiscal 2026 filing; that disclosure is specific to the issuer, while the underlying questions are useful to apply to any target. See C3.ai’s risk disclosures.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →4. Follow the costs of delivering AI
For each major product or segment, identify the costs that may change as use grows. Depending on the business, they can include model inference, GPUs or other compute, cloud hosting, data rights or licensing, human review, implementation, support, and research and development. Verify the company’s actual disclosures rather than inferring its cost structure solely from the fact that it uses AI.
Compare these costs with revenue per customer or unit of use. Determine whether customers pay separately for heavy usage, whether a subscription includes usage that can vary widely, and whether the company can pass through higher costs. Also assess whether efficiency improvements, model routing, or infrastructure changes could reduce the cost to serve without harming product quality.
A 2026 SEC-filed AI infrastructure issuer identifies possible risks including volatile usage-based revenue, subscription pricing that may not capture heavy usage, pricing below inference costs, and commoditization pressure on prices and gross margins. These are mechanisms to investigate in a target’s own disclosures—not findings that apply to every AI company. Read GridAI Technologies Corp.’s 2025 Form 10-K.
5. Connect growth to margins, cash, and funding
Review several reporting periods together. Revenue growth can coexist with deteriorating margins, heavy operating losses, or cash consumption. Examine gross profit and gross-margin trends, operating expenses, operating cash flow, capital expenditure, stock-based compensation, cash balance, debt, and any need for additional financing. Read management’s explanation of what must improve before the business can fund itself, and compare targets or projections with results already achieved.
C3.ai’s fiscal 2025 Form 10-K reported net losses of $288.7 million in fiscal 2025, $279.7 million in fiscal 2024, and $268.8 million in fiscal 2023, and an accumulated deficit of $1.4 billion as of April 30, 2025. These are historical C3.ai figures, not current results or a claim about AI companies generally. They illustrate why an investor should read growth alongside profitability and funding information. Read C3.ai’s fiscal 2025 Form 10-K.
For a private company, public disclosures may not provide enough information to evaluate unit economics, churn, cash runway, or customer concentration. Mark missing metrics as unknown; do not treat an absence of disclosure as evidence of strong performance.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.6. Assess what could protect the business—and what could break it
Test whether customers have a reason to keep paying beyond the current model or feature set. Possible sources of durability include demonstrated customer outcomes, rights to valuable data, deep workflow integration, distribution, reliability, and scale. For each claimed advantage, ask whether the company owns or controls it, whether customers value it, and whether a competitor could reproduce it or bundle a substitute.
Review dependence on model, chip, cloud, and data suppliers, alongside privacy, security, intellectual-property, regulatory, and execution risks. Microsoft’s fiscal 2025 Form 10-K describes significant AI development and operating costs and a rapidly evolving, competitive market. That supports treating competition and cost as live diligence topics, but target-specific exposure must be assessed from the target’s own filings. Read Microsoft’s fiscal 2025 Form 10-K.
Best Value
Write down a few concrete downside cases and trace their effects through revenue, margin, and cash:
- Customers delay production adoption after pilots, slowing or deferring expected revenue.
- Inference or hosting costs rise faster than the company can improve efficiency or adjust prices.
- A major customer does not renew or reduces usage.
- A model provider or cloud platform bundles a competing feature, or lower-cost and open alternatives weaken pricing power.
- A reliability failure, privacy incident, legal restriction, or execution problem raises costs or reduces demand.
Turn the filing review into an investment screen
Before reaching a conclusion, write down what the company sells, who pays, how revenue recurs, what it costs to serve customers, and what evidence supports continued demand. For each answer, distinguish reported results from management forecasts and note the relevant period. If a key answer depends on information the company has not disclosed, keep it marked as unknown rather than filling the gap with an industry assumption.
A sound business model does not, on its own, establish that a stock is attractively priced. Valuation, dilution, governance, the investor’s time horizon, and tolerance for risk require separate analysis; this framework is a way to examine the business, not a buy or sell recommendation.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
