Free tools Windows power users keep installed
One-click scans. No signup required.
Before buying an AI stock, verify what the company actually sells, whether customers are paying for it repeatedly, how much cash it consumes to deliver the product, and whether its financing, competitive position and legal risks can withstand slower growth. An AI feature, a large addressable market and a rising share price are not evidence of durable profits.
Use the latest annual and quarterly filings, then reconcile management’s narrative with revenue, margins, cash flow, customer behavior, infrastructure commitments and dilution. The framework below is designed for a named public issuer; date every financial and valuation figure because AI businesses and market prices change quickly.
As an Amazon Associate I earn from qualifying purchases.
First, identify what kind of AI company you are analyzing
“AI company” can describe businesses with very different economics. Classify the issuer before interpreting its growth or valuation.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Pure-play AI software or model vendor
A pure-play vendor’s revenue may depend on subscriptions, usage-based charges, contracts for model access, implementation services or a combination. Its upside can be concentrated, but so are risks: a few customers may represent a large share of sales, compute costs can rise with usage, and a better-funded competitor or open-source alternative can pressure prices.
#1 Best Overall
Diversified technology company with AI products
A diversified company may sell cloud, software, devices or services while investing heavily in AI. AI revenue can be difficult to isolate from the consolidated results. The company may have more funding and distribution, but its AI investment can still reduce margins or returns on capital if demand, pricing or adoption disappoints.
Company using AI in its own operations
A bank, retailer, manufacturer or software company can use AI to reduce costs or improve products without selling AI as a separate product. The investment case rests on measurable productivity, retention or cross-selling benefits, not on a standalone AI revenue line. Ask how management measures those benefits and when they should appear in reported results.
| Business type | Revenue question | Cost question | Key evidence |
|---|---|---|---|
| Pure-play vendor | Are recurring or usage revenues growing, and are customers renewing or expanding? | Do compute, data, implementation and support costs scale more slowly than revenue? | Revenue mix, retention, gross margin, customer concentration and cash flow |
| Diversified provider | Which products and segments actually contain AI revenue? | How much capital expenditure, research and infrastructure spending is allocated to AI? | Segment disclosures, capital-expenditure plans and management’s return targets |
| AI-enabled operator | Is there evidence of higher sales, retention or productivity attributable to AI? | What implementation, labor, data and compliance costs are required? | Operational metrics, margins and controlled descriptions of efficiency gains |
Read the latest filings before reading forecasts
Start with the issuer’s latest Form 10-K and most recent Form 10-Q. Read the business description, management’s discussion and analysis, financial statements, notes, contractual obligations, liquidity discussion and risk factors. Then check earnings releases and investor presentations against the filed numbers.
A risk-factor summary is not a substitute for the full filing. Veritone’s FY2025 Form 10-K explicitly says its summary does not contain all information that may be important and directs readers to the detailed risks and the rest of the annual report. Read the filing at the SEC.
- Confirm the period and filing date. Record the fiscal year-end, quarter-end and filing date. Do not mix a current market price with an old financial period without labeling the mismatch.
- Read the business and segment descriptions. Identify products, customer types, geographic exposure, contract terms and whether AI is a product, an infrastructure layer or an internal tool.
- Trace management’s explanations to statements and notes. If management says demand is accelerating, look for corresponding billings, remaining performance obligations, deferred revenue, customer additions or usage data.
- Review every risk category. Note risks that could affect revenue, costs, capital access, intellectual property, privacy, security, regulation and leadership.
- Check subsequent events and guidance. A filing can become stale after quarter-end; later filings and current disclosures may change the interpretation.
For a concrete example of issuer-specific disclosures, C3.ai’s FY2026 Form 10-K is available from the SEC, and Microsoft’s FY2026 Form 10-K is available from the SEC. These filings describe different business models; neither is a universal template for the sector.
Test whether revenue is real, repeatable and profitable
Revenue growth matters only when the underlying demand is durable and economically attractive. Separate reported growth from the quality of that growth.
Rank #2
- Ideal for Gifting
- Ideal for a bookworm
- Compact for travelling
Check the source of each dollar
- Break revenue into subscriptions, consumption or usage, licenses, services, hardware and one-time contracts.
- Determine whether a large contract is recognized immediately or over its service period.
- Compare reported revenue with billings, deferred revenue, remaining performance obligations and cash collections when disclosed.
- Ask whether growth comes from more customers, larger customers, higher usage, price increases, acquisitions or favorable foreign-exchange movements.
Usage-based revenue can rise quickly when customers experiment and fall just as quickly when projects are stopped. Subscription revenue can look steadier but still face nonrenewal, downgrades and discounting. Professional services may help deploy a product while producing lower margins than software.
Recommended Free Tools
Measure customer demand beyond logos
Count paying customers rather than press-release logos, and look for renewal, expansion and concentration data. A company that adds customers but loses existing accounts may be buying growth. Ask how many customers generate a material share of revenue, whether one contract is unusually large, and whether the company discloses annual recurring revenue, net retention or backlog consistently enough to analyze.
Separate growth from sustainable profitability
Track gross margin, research and development, sales and marketing, general and administrative expense, stock-based compensation, operating cash flow and free cash flow. Then read management’s explanation of what must change for sustainable profitability: pricing, utilization, lower compute cost, slower hiring, a different product mix or simply much greater scale.
C3.ai reported net losses of approximately $470.4 million in FY2026, $288.7 million in FY2025 and $279.7 million in FY2024. Those are company-reported figures for C3.ai, not an AI-industry benchmark. Its FY2026 filing also reports an accumulated deficit of approximately $1.8 billion as of April 30, 2026. The filing states: “We have a history of losses, we anticipate our operating expenses will continue to increase, and we may not be able to achieve or maintain profitability in the future.” Read the company’s full context rather than treating the figures as a forecast.
| Metric | What to calculate or compare | What a deterioration can indicate |
|---|---|---|
| Revenue growth | Year-over-year and sequential growth by product and segment | Slowing demand, contract timing or a maturing market |
| Gross margin | Revenue less hosting, infrastructure, support and other direct costs | Compute inflation, discounting, services mix or poor utilization |
| Operating expenses | R&D, sales, marketing and administration as a share of revenue | Growth purchased through spending without operating leverage |
| Cash flow | Operating cash flow and capital expenditure relative to revenue | Accounting growth that is not converting to cash |
| Customer economics | Retention, expansion, acquisition cost and payback where disclosed | Customers that are expensive to win or do not renew |
Calculate the AI cost and return equation
AI products can require substantial spending before revenue arrives. Analyze the complete cost stack, not only reported research and development.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Research and model development
Review R&D expense, employee growth, specialized compensation and the company’s description of model training or product development. Capitalized software costs, if any, should be read in the accounting notes so you understand which costs are expensed and which are recorded as assets.
Compute, data centers and energy
Ask whether the issuer owns infrastructure, leases capacity, buys cloud services or uses a hybrid model. Examine capital expenditure, purchase commitments, cloud-contract obligations, depreciation and energy exposure. A diversified provider’s infrastructure scale and financing capacity may differ sharply from a smaller software specialist’s; do not transfer one company’s cost structure to another.
Deployment and support
Implementation consultants, data preparation, model monitoring, customer support, cybersecurity and compliance can consume people and cash after a contract is signed. Compare services revenue and services cost with software margins. If each deployment requires substantial customization, growth may require proportional hiring rather than producing software-like operating leverage.
Microsoft’s FY2026 filing describes substantial AI and cloud investment and risks involving evolving demand, costs, pricing, customer adoption and returns. Those disclosures demonstrate the questions to ask, not a claim that every AI issuer faces the same scale or capital burden. Read the filing here.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Turn spending into a return test
- Estimate the revenue attributable to the AI product or use case, using the company’s stated definition.
- Subtract direct compute, hosting, data, support and delivery costs to estimate contribution margin.
- Subtract incremental R&D, sales and compliance spending needed to sustain growth.
- Compare the resulting cash contribution with capital invested and the time required to reach it.
- Stress-test the result for lower usage, price reductions, higher inference costs and delayed customer adoption.
Assess whether the competitive position can last
Technology leadership is temporary unless customers have a durable reason to stay. Identify the specific advantage and the mechanism that protects it.
Ask why customers choose the product
Possible advantages include proprietary data rights, workflow integration, distribution, switching costs, reliability, security certifications, domain expertise or lower total cost. “Uses advanced AI” is a capability description, not a moat. Verify whether customers can replace the product with an internal system, a cloud-provider service, an open-source model or a lower-priced rival.
Look for evidence of pricing power
Stable or rising gross margins, renewals without heavy discounting and expansion within existing accounts are stronger evidence than a large pipeline. Read disclosures for price competition, customer procurement pressure and the possibility that model capabilities become widely available.
Rank #4
Microsoft describes AI as a highly competitive, rapidly evolving market. Veritone says the market for AI-based software applications is relatively new and unproven and may decline or experience limited growth. These are company risk disclosures, not predictions that all AI companies will fail or succeed. Compare each issuer’s alternatives, customers and product maturity.
Map concentration and dependency
Check dependence on one cloud provider, chip supplier, data source, channel partner or large customer. A favorable partnership can accelerate distribution while also creating bargaining and continuity risk. Read termination rights, minimum commitments and exclusivity provisions in the notes when disclosed.
Check liquidity, debt and dilution before trusting the runway
A company can report strong bookings and still need frequent financing. Build a simple liquidity schedule from cash and equivalents, restricted cash, marketable securities, operating cash burn, capital expenditure, debt maturities, interest expense and committed purchases.
Estimate runway using the issuer’s own cash use
Use recent operating cash flow, expected capital expenditure and management’s spending outlook. Present a range rather than a single precise month count because growth investment and working capital can change quickly. A cash balance alone does not show how much is available after restrictions, customer obligations or debt covenants.
Inspect dilution
- Read the share-count reconciliation and basic versus diluted shares.
- Review stock-based compensation, option overhang, restricted shares and employee purchase plans.
- Check at-the-market programs, shelf registrations, convertibles, warrants and recent equity offerings.
- Compare per-share growth with total-company growth; issuing shares can reduce each existing holder’s percentage ownership.
Veritone identifies indebtedness and liquidity among its risk categories, while C3.ai describes continuing losses and anticipated expense increases. Use those filings as prompts, not as estimates of another issuer’s runway.
Read risks that do not appear in the product demo
Intellectual property and data rights
Determine who owns training data, customer inputs, model outputs, fine-tuned weights and software code. Look for indemnities, infringement claims, licensing restrictions and provisions that shift liability to the company. A product can be technically impressive yet commercially constrained if its data rights are uncertain.
Best Value
- It can be a gift option
- Comes with secure packaging
- Helpful in various ways
Privacy, cybersecurity and misuse
Review security incidents, data handling, access controls, model monitoring and contractual promises. AI systems can expose confidential information, generate harmful content or be misused by customers. Remediation, notification and litigation costs can be material even when no revenue is immediately lost.
Regulation and public scrutiny
Identify the jurisdictions, regulated customers and use cases affected by privacy, consumer-protection, employment, copyright, export or sector-specific rules. Microsoft discusses AI misuse, regulation and public scrutiny in its FY2026 filing. The relevant exposure for a particular issuer depends on its role as model developer, software provider, deployer or user.
Execution, leadership and governance
Read related-party transactions, auditor changes, internal-control disclosures, executive turnover, incentive targets and board expertise. Compare promised milestones with later filings. Frequent strategy changes, unexplained non-GAAP adjustments or aggressive customer and pipeline claims warrant additional skepticism.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsOnly after the business test, analyze valuation
Valuation should connect the share price to the company’s actual growth, margins, cash requirements and risks. A high growth rate does not establish fair value, and promising technology does not establish that a stock suits your objectives or risk tolerance.
Use comparable companies carefully
Compare businesses with similar revenue models, customer types, margins, capital intensity and accounting periods. A profitable diversified provider should not be valued with the same metric as a loss-making early-stage vendor. Label whether you are using enterprise value to revenue, price to sales, enterprise value to gross profit, price to earnings or a cash-flow measure, and keep numerator and denominator periods consistent.
Build a scenario range
| Scenario | Operating assumptions to test | Questions for the stock |
|---|---|---|
| Upside | Strong customer adoption, improving gross margin and disciplined infrastructure spending | Does the current price already assume this outcome? |
| Base | Moderating growth, continued investment and a gradual path toward break-even | Can the company fund the plan without substantial dilution? |
| Downside | Slower usage, price pressure, higher compute costs or regulatory delays | What cash, covenant or competitive trigger would force new financing? |
No current valuation conclusion is supported here because no issuer, market price or comparable valuation set was specified. Perform the calculation only after choosing a company and recording the date, share count, net debt, forecast period and assumptions.
A practical pre-purchase checklist
- Name the issuer and date your work. Record ticker, exchange, fiscal year-end, latest filing date and stock-price date.
- Classify the business. Mark it as pure-play, diversified provider or AI-enabled operator, and identify the products that generate the claimed AI exposure.
- Read the full 10-K and latest 10-Q. Capture revenue mix, customer concentration, margins, cash flow, debt, commitments and every material risk.
- Reconcile demand claims. Check paying customers, retention, usage, backlog, billings, deferred revenue and collections where disclosed.
- Model unit economics. Estimate direct compute, data, delivery and support costs before drawing conclusions from gross margin.
- Stress-test funding. Calculate cash needs under slower growth, higher infrastructure costs and continued operating losses; include dilution.
- Map alternatives. List internal builds, open-source options, incumbent platforms and likely substitutes, then identify the company’s switching costs.
- Review legal and governance exposure. Check data rights, intellectual property, privacy, cybersecurity, regulation, leadership and controls.
- Compare valuation last. Use genuinely comparable companies, consistent periods and explicit scenarios.
- Match the risk to your plan. Decide how much loss, volatility, dilution and thesis uncertainty your portfolio can tolerate before placing an order.
Red flags that should slow a purchase
- Revenue growth depends on one unusually large contract or an undisclosed customer concentration.
- Reported usage rises while cash collections, retention or gross margin weaken.
- Management emphasizes bookings or adjusted metrics while omitting cash burn and stock compensation.
- Infrastructure commitments grow faster than credible demand, or the company cannot explain expected returns.
- Repeated equity issuance, convertible debt or option grants materially increase the future share count.
- Model performance is highlighted without evidence of production deployment, renewal or customer economics.
- Risk factors describe unproven markets, intense competition, regulatory uncertainty or intellectual-property exposure that the valuation appears to ignore.
The decision standard is simple: an AI narrative must survive a filing-based test of demand, economics, funding, competition and risk. If one essential input is not disclosed, treat the uncertainty as part of the investment case rather than filling the gap with assumptions.
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.
