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 →Before investing in an AI company, ask for evidence that customers pay for a real solution, that the product works in its intended setting, and that the business can earn money without unacceptable legal, technical or financing risks. Then examine what you are buying: public shares and private offerings have different disclosures, rights and liquidity. Treat demonstrations, forecasts and AI branding as claims to verify—not proof of investment quality.
Who has the problem, and who pays to solve it?
Start with the customer and the transaction, not the model. Ask the company to identify its target buyer, the problem the product solves, how it fits into the buyer’s workflow, and who approves the purchase. Then distinguish evidence of a durable business from evidence that people are merely interested.
As an Amazon Associate I earn from qualifying purchases.
- How many customers are paying for the product, and how many are still in pilots or trials?
- Do customers renew, expand usage or add seats? What do retention and expansion look like over time?
- How long does implementation take, and how soon does a customer receive value?
- What measurable result does the product deliver—such as time saved, fewer errors or lower operating costs—and how is that result calculated?
- Can customers independently confirm the results and explain what would make them stop using the product?
A launch announcement or successful demo does not establish sustained paid use. Prefer customer evidence that is specific, repeatable and relevant to the product’s actual deployment. For public issuers, examine how management describes adoption and customer value in its filings; for a private company, request deployment, renewal and customer-reference evidence appropriate to its stage.
What does the AI actually do, and how well does it work?
Ask which parts of the product rely on AI and what happens when those parts are wrong, incomplete or unavailable. A polished demonstration shows a possible outcome, not necessarily typical performance in a customer’s working environment.
#1 Best Overall
- What tasks does the system perform, and which tasks still require a person?
- What evaluations or benchmarks support the performance claims? Do they reflect real customer data, workflows and edge cases?
- How often does the system produce inaccurate, misleading or unusable outputs in the intended setting?
- How are errors detected, corrected and escalated? What safeguards prevent an incorrect output from triggering a consequential action?
- How does the company monitor performance after deployment, including changes in customer data, model versions or usage?
Ask for limitations and failure examples alongside success cases. Determine whether human review is a temporary launch measure or a continuing operating cost, and whether the customer—not the vendor—must absorb that work. Inaccurate outputs, model defects and inadequate oversight can create business, legal and reputational harm; the significance depends on the product and how customers use it.
Does the company have the rights and controls for its data?
Data can be a product input, a customer asset and a source of legal or security exposure. Ask the company to explain the origin and permitted use of the data it collects, uses to train or improve systems, retains, and provides to customers.
- What data is used for training, fine-tuning, retrieval or evaluation, and what rights permit each use?
- Can customer data be used to improve shared models, or is it kept separate? What do the customer contracts say?
- How long is information retained, where is it processed, and how can a customer request deletion?
- What security controls protect customer data, and how does the company respond to a breach or other incident?
- Which third-party models, datasets, cloud services or compute providers are essential? What happens if their price, availability or terms change?
Look for documented rights and controls, not a general assurance that data is “proprietary” or “secure.” Poor-quality or biased data, unclear permissions, privacy or contract violations, cybersecurity threats and weak governance can all undermine a product or expose the company to disputes.
Can the business make money as usage grows?
Revenue alone does not show whether an AI product has attractive economics. Examine the cost of delivering the service—including compute and inference—as well as research, support, compliance and sales costs. Ask whether heavier customer use improves the economics, leaves them roughly unchanged or makes them worse.
- What are gross margins after infrastructure and other direct service costs, and how have they changed as usage has grown?
- How concentrated is revenue among the largest customers? What would happen if one did not renew?
- How much cash is the company using, what drives that spending, and when might it need more financing?
- Are renewals and expansions sufficient to support the company’s growth plans?
- What evidence supports management’s projections, and which assumptions would most alter them?
Review audited or otherwise reliable financial statements when available, and compare forecasts with actual results. Investor.gov’s private-placement guidance recommends reviewing financial statements, assessing whether claims and expectations are reasonable, and understanding how proceeds will be used. Its guidance is general investor education, not a substitute for reviewing the particular offering documents: Private Placements under Regulation D – Updated Investor Bulletin.
What makes the product hard to replace?
Compare the company with direct competitors, established software vendors, in-house solutions, open models and the providers on which the company itself depends. An AI label or a technically impressive model does not by itself establish a durable advantage.
- What would a customer lose by switching to another vendor or building an alternative?
- Does the product have durable access to unique data or workflows, and are the rights to that access clear?
- Can a customer substitute a general-purpose or open model, or add similar features to software it already uses?
- Could the company’s model, compute or platform suppliers compete with it, raise prices or change terms?
- What could make the product less useful or less differentiated as technology and customer needs change?
Look for a defensible reason customers will continue to choose the product, supported by customer behavior and lawful access to the resources it needs. Investor.gov advises investors to identify competitors; the relevant alternatives may extend beyond companies that market themselves as AI businesses.
What legal, security and governance obligations apply?
Legal exposure depends on what the product does, where it operates and who uses it. A tool that assists a low-impact internal task presents different questions from one whose outputs affect people’s access to important services or decisions. Ask how the company identifies those obligations and assigns responsibility.
- Which countries, states or other jurisdictions does the company serve, and which customer industries does it target?
- Who reviews consequential outputs, handles complaints and escalates incidents?
- How are privacy, intellectual-property and security obligations addressed in product design and customer contracts?
- Who bears responsibility if the system causes harm or violates a customer’s requirements? What do indemnities and insurance cover?
- Does the company have people and processes capable of keeping up with applicable rules as products and markets change?
Do not assume a single answer applies to every AI product or jurisdiction. Have qualified counsel assess the particular company, use case and transaction. Potential exposures include litigation, regulatory scrutiny, data-rights disputes, security incidents and reputational damage.
What are you buying, and what could happen to your ownership?
Read the actual security and transaction documents. A belief in the company’s product is not enough if the investment terms, dilution risk or route to liquidity are unsuitable. For public shares, start with the latest filings and share structure. For a private offering, review the offering documents and understand the precise instrument and investor rights.
| Question | Public shares | Private offering |
|---|---|---|
| What documents should you examine? | Latest filings, financial statements, share structure and risk factors. | Offering documents, financial statements where available, security terms and use of proceeds. |
| What can change your ownership or economics? | Share issuance, dilution, cash needs and changes in the business. | Valuation basis, fees, conversion or liquidation terms, future financing and dilution. |
| What rights and liquidity should you understand? | Share class and rights, plus the practical ability to sell shares. | Voting and information rights, transfer restrictions, and whether an exit is realistic or may take a long time. |
For a private investment, ask what the security converts into, what happens in a sale or liquidation, and whether fees or other terms affect your return. Understand the offering exemption and the limits of the documents you receive. Investor.gov warns that private placements can be highly illiquid, may provide less disclosure and can result in a total loss. A Form D notice is not SEC approval or registration; its presence should not be presented as an endorsement of the company or offering. The bulletin also advises: “If an issuer fails to adequately answer your questions, consider this a warning against making the investment.” That is Investor.gov guidance, not a binding SEC rule.
Free tools Windows power users keep installed
One-click scans. No signup required.
Who is making the claims, and what incentives do they have?
Verify who is behind the investment and how they benefit if you invest. Check management backgrounds and references, independently corroborate material claims, and understand how promoters, representatives or intermediaries are compensated. Ask follow-up questions until you understand the product, the financial picture and the security you would own.
Best Value
Pause if answers are evasive, important information is missing, or you are pressed to commit before you can review the documents. Guaranteed returns, implausible claims and suggestions that SEC filing or a prominent backer makes an investment safe warrant particular scrutiny. Investor.gov’s private-placement bulletin explains that Form D is a notice filing, not SEC approval, and discusses risks and warning signs for investors.
How should you compare AI companies?
Use the same evidence standard for each company rather than giving one a pass because it has a familiar name, a prominent investor or an “AI-first” description. A concise comparison should cover:
- Paid adoption, renewals, expansion, customer concentration and independently verifiable outcomes.
- Performance in real workflows, failure modes, human oversight and time to value.
- Data and intellectual-property rights, privacy and security controls, and dependence on outside models or compute.
- Gross margin after AI infrastructure costs, cash burn, capital needs and financing risk.
- Differentiation, switching costs, alternatives and resilience to technology changes.
- Legal and governance exposure for the product’s use cases and operating geographies.
- Valuation, security rights, dilution, fees, liquidity and plausible exit paths.
Private-company information may be limited, and its valuation can be harder to assess than that of listed shares. If a company cannot provide evidence for a material claim, record that as an unanswered question rather than treating the claim as established.
Recommended Free Tools
When should you stop and verify before investing?
Pause the decision if the company cannot explain who pays and why, offers demonstrations without credible deployment evidence, cannot substantiate its performance or data rights, obscures customer concentration or funding needs, relies on unsupported forecasts, or will not explain how the security works. In a private offering, urgency, promises of guaranteed returns, missing offering information or claims of SEC approval deserve particular scrutiny.
This framework cannot determine whether a specific AI company is a good investment. That requires current company filings or offering documents, corroboration of material claims, a clear understanding of the security and transaction terms, and advice suited to your jurisdiction and circumstances. Risk disclosures from individual issuers illustrate possible issues; they do not establish that every AI company faces the same risks.
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.
