Lawmakers in 30+ States Are Eyeing AI Pricing. Here’s What a Small Business Should Check Before Using It.

Consumers are getting angry and confused about how prices are set. According to an AP report this week, shoppers worry that digital price tags and AI tools mean different people pay different prices for the same thing, and lawmakers are paying attention. The National Conference of State Legislatures says legislators in at least 30 states have considered pricing bills this year. The FTC has proposed a policy that targets undisclosed personalized pricing. You are probably not running a grocery chain, but if you use AI to set prices, build quotes, run discounts, or decide who sees which offer, this is now your issue too.

What Is Actually Happening

Retailers and restaurant chains are adopting software that predicts how customers will respond to a price. A Simon-Kucher survey of more than 2,200 business leaders found that 41 percent used AI to gather market intelligence, including competitors’ prices, and 28 percent used it for real-time price optimization. Simon-Kucher is a pricing consultancy, so treat the numbers as directional.

Lawmakers and regulators are responding. According to reporting on the NCSL data, Maryland and New Jersey have enacted restrictions on grocery pricing, and Connecticut approved broader protections for retail purchases. The FTC has proposed an enforcement policy warning that failing to disclose the use of personal data to set individualized prices could violate consumer protection law. It is a proposal, not a final rule. A consumer lawsuit against McDonald’s alleges its pricing system uses nonpublic sales data from competing locations to coordinate prices. McDonald’s says the suit is filled with inaccuracies and that franchisees use the tool to set their own prices. A federal case over Amazon’s pricing tool is scheduled for trial next year, and Amazon disputes the allegations.

The Honest Counterpoint

None of this proves AI pricing is hurting shoppers. A researcher tracking prices at one grocery chain found virtually no surge pricing after electronic shelf labels arrived. A pricing-data company says about half the price changes it recorded were decreases. Both sources have a stake in the answer, so they are not neutral. The real picture is that the technology is mostly used for ordinary price management today, while the fear is about where it could go.

That distinction is why this matters for you. The risk is not that your tool is doing something wrong. It is that you cannot easily show a customer or a regulator that it is not.

Why a Small Business Should Care

Large companies have legal teams to answer these questions. Small businesses usually have a vendor’s settings page and a hunch. If you use software that adjusts prices by demand, quotes jobs automatically, or personalizes discounts by customer, you are the one who has to explain it when someone asks. A customer who learns a neighbor got a lower quote for the same work will not care how the algorithm works. They will care whether it feels fair.

Five Questions to Ask Before You Use AI on Pricing

First, what data goes in? Competitor prices, season, and demand are one thing. A customer’s location, browsing history, or past purchases are another, and that second category is what the FTC’s proposed policy is about.

Second, could two customers get different prices for the same thing? Different prices for different services, volumes, or dates are easy to explain. Differences that trace back to who the customer is are where the risk sits.

Third, do you tell people? If personal data influences a price or an offer, say so in plain language. Silence is the problem the FTC’s proposal describes.

Fourth, who approves changes? Put a human name on the pricing rules, and set floors and ceilings so a tool cannot swing prices overnight. This is the same permission list idea we have covered for agents, applied to money.

Fifth, can you reconstruct what happened? Keep a record of what price was shown, to whom or under what rule, when, and why. If a customer or a regulator ever asks, a log is the difference between an explanation and a guess.

The Honest Takeaway

This is not legal advice, and the rules vary by state and are still changing, so talk to counsel about your own setup. But the principle holds up whatever the rules end up being. Set prices based on the job, not on who is asking, and be able to explain how you got there in one sentence. Trust is what customers pay for, and a price they feel was fair is worth more than a few extra dollars from a tool.

---

Also read: Young Americans Don’t Trust AI CEOs. Here’s What That Means If AI Is Part of Your Pitch.