Consider the moment you do not see. It is an ordinary Tuesday evening, and you open a grocery app to order the week's essentials. Nappies, milk, bread, the brand of coffee you always buy, the painkillers a household runs through unnoticed. You add the items, glance at the total, tap to confirm. The total seems about right. You have nothing to compare it against, because there is nothing to compare it against. The price you see is the only price you will ever see. You do not know, and have no way of finding out, that the shopper in the next postcode, ordering the identical basket from the identical store at the identical minute, has been quoted a figure several pounds lower. You do not know that a piece of software has looked at what it can infer about you, your past behaviour, your location, the predictability of your needs, the apparent absence of alternatives, and concluded that you, specifically, will pay a little more. No negotiation, no notice. There was only a number, presented as if it were the number, and you accepted it because the entire architecture of shopping has trained you to assume that a price is a fact about a product rather than a judgement about you.
This is not a thought experiment. In December 2025, a joint investigation by Consumer Reports, the Groundwork Collaborative and More Perfect Union pulled back the curtain on exactly this practice, running inside Instacart, the largest grocery delivery platform in the United States. The investigation found that roughly three-quarters of products checked were being offered to different customers at different prices, for the same item, from the same store, at the same time. The variations ran from a few pennies to more than two dollars per item. Extrapolated across a typical household's annual spend, the swing came to around 1,200 dollars a year. The engine behind it was an artificial intelligence pricing platform called Eversight, which Instacart had acquired in 2022, and which the company marketed to retailers as a way to lift sales and squeeze out incremental margin. Within days of the story being published, Instacart announced that, effective immediately, it was ending all item price tests on its platform. The lab, as one campaigner put it, had been closed only because someone finally switched on the lights.
The episode is not an aberration. It is a preview. The capacity to set a different price for every customer, calibrated to the maximum each will tolerate, has been the holy grail of commerce for as long as commerce has existed, and for almost all of that history it has been impossible at scale. What has changed is that the impossibility has dissolved. Cheap data, behavioural tracking and machine learning have made it not merely feasible but routine to estimate, in real time, how much a particular human being is likely to pay, and to charge them precisely that. The question this raises is not technical. The technology works. It is what it means to live in a market where the price is no longer a shared fact about the world but a private message addressed to you alone, written in a language you cannot read, by a system that knows things about you that you have not agreed to disclose and may not even know yourself.
The Oldest Dream in Retail
Economists have a name for what Instacart's software was reaching towards, and it is not new. They call it first-degree price discrimination, or perfect price discrimination, and it describes the seller's fantasy of charging each buyer exactly their maximum willingness to pay. The market trader who sizes up a customer's shoes and accent before naming a figure is practising a crude, intuitive version of it. The theory has been in textbooks for a century. What it has lacked, until very recently, is a mechanism. To charge everyone their personal maximum, a seller must somehow know everyone's personal maximum, and individual human beings have historically been quite good at concealing it. The posted price, the same number on the same shelf for everyone, emerged in part because sellers could not do better. It was a technological limit dressed up as a social norm.
The first sign that the limit might be lifting came in September 2000, when shoppers on Amazon noticed something strange. A man buying a DVD found that when he deleted the cookies from his browser, the price dropped. Amazon, it turned out, had been running an experiment in which the price of certain titles varied according to what the company could infer about the shopper from their browsing and purchase history. Loyal customers, the kind least likely to wander off, were in some cases being shown higher prices than newcomers. The discovery produced a wave of public fury, and Amazon retreated almost at once, insisting the variations had been a random test rather than deliberate profiling, and refunding the difference. The episode entered the folklore of e-commerce. The lesson the industry drew was not that personalised pricing was wrong. The lesson was that it must never again be visible.
For the better part of two decades the dream advanced quietly, in forms ordinary shoppers had been trained to accept. Airlines pioneered the art, charging fares that lurched with demand, with the day of the week, with how close the departure loomed, and, as many travellers suspected, with how many times a route had been searched from a given device. Ride-hailing apps normalised the idea that a price could surge in real time, rising when it rained or when a concert let out, framed as a neutral response to supply and demand rather than a calculation about the rider's desperation. Streaming services and online retailers learned to offer a discount to one customer that never materialised for another. Each of these was a step away from the posted price and towards the personalised one, and each was small enough, and dressed in enough economic respectability, that it provoked little sustained alarm. The frog, to borrow the old image, was being warmed by degrees.
What the Machine Sees
The leap from dynamic pricing, where the figure moves with the market, to surveillance pricing, where the figure moves with the customer, is a leap in kind and not merely degree. A surge fare is at least the same for everyone standing on the same wet pavement at the same moment. Surveillance pricing is the surge fare turned inward, aimed not at the conditions but at the person. The raw material it runs on is the vast, largely invisible economy of behavioural data that has accreted around every digital interaction we have.
In January 2025, the United States Federal Trade Commission published the initial findings of a study into precisely this market. Acting under its Section 6(b) authority, which lets it compel companies to hand over internal documents, the agency had sent orders to a clutch of intermediaries that sit, mostly unseen, between retailers and shoppers: Mastercard, Accenture, the pricing-software firm PROS, the personalisation company Bloomreach, the pricing optimiser Revionics and the consultancy McKinsey and Company. What the staff found, even in a preliminary cut, was a thriving and shadowy infrastructure for setting individualised prices. The intermediaries drew on a remarkable breadth of signals, both data volunteered by consumers and data inferred about them from first and third party sources. The behaviours that could be tracked and fed into a price ranged from the movements of a mouse across a webpage to the specific products a shopper abandoned, unpurchased, in an online basket. One example in the documents was a cosmetics company targeting promotions by a customer's skin type and skin tone. The intermediaries the FTC examined were, between them, working with at least 250 clients selling everything from groceries to clothing. The market for knowing what you will pay was already industrial in scale.
The Instacart investigation gave that abstraction a face. When Consumer Reports and its partners examined the patent filings that Instacart and Eversight had lodged from 2017 onward, they found the ambition spelled out in the dry language of intellectual property. The patents referenced setting prices using previous purchase history, buying behaviour, and characteristics such as age, gender, household size and household income. One metric flagged was whether a shopper was new to a brand or returning to it. The investigation also documented what it called phantom discounts, in which different customers were shown different inflated original prices for the same item, manufacturing the impression of a bargain where none existed. A box of premium saltine crackers, in one example, was presented with an original price of 5.93 dollars, 5.99 dollars or 6.69 dollars depending on the shopper, before a sale price of 3.99 dollars that was in fact the same for everyone. The discount was theatre. The variation was real.
Instacart denied that it currently used personal or demographic data to set prices, maintaining that customers were randomly assigned to pricing cohorts by product category and location rather than profiled as individuals. But the denial, even taken at face value, missed the point the industry's own analysts kept returning to. Phil Lempert, a grocery analyst who runs the site SupermarketGuru, put it plainly: once the technology is in place, even if a company is not profiling shoppers today, the capacity to start is a button-press away. The machinery of individualised pricing does not need to be aimed at you to be pointed in your direction. Its mere existence changes the relationship between buyer and seller, removing the floor of the posted price and replacing it with an open question about how much, in your case, the seller thinks it can get.
The Survey Nobody in Retail Wanted to Read
Defenders of personalised pricing tend to argue that consumers do not really mind, or that they accept it as the price of convenience, or that the discounts it enables for the price-sensitive outweigh the premiums it imposes on the rest. The data does not support this. As part of its investigation, Consumer Reports ran a nationally representative survey of 2,240 American adults in September 2025. Among those who had used Instacart in the previous year, 72 per cent did not want the company to charge different users different prices for any reason. Not for some reasons. Not unless the reasons were fair. For any reason at all. The aversion was close to universal, and it cut against the entire logic of the surveillance-pricing business.
This exposes the gap between what the practice does and what it claims to do. The economic defence of first-degree price discrimination holds that it can, in theory, expand the market, letting sellers profitably reach price-sensitive buyers who would otherwise be excluded while extracting more from those who can afford it. On a whiteboard this looks almost progressive, a kind of automated means-testing. In the world it works the other way around. The signals a machine-learning system finds most useful for estimating willingness to pay are precisely the signals that track vulnerability. A shopper in a food desert, with no rival supermarket within reach, has fewer alternatives, and the algorithm can learn to read that constraint and charge for it. A household ordering nappies and prescription items has predictable, inelastic needs, and inelasticity is exactly what a pricing model is built to exploit. The customer with limited mobility, least able to drive between shops, is least able to escape and therefore most worth charging more. The system does not optimise for fairness. It optimises for revenue, and the people with the least room to push back are the ones from whom there is the most to extract.
Lina Khan, who chaired the FTC from 2021 to 2025 and now teaches at Columbia Law School, framed the stakes in a sentence that has stuck to the debate. We are moving, she said, from a transparent market with public prices to an opaque world where we are alone against secret algorithms. The phrasing identifies the precise thing that is lost. It is not simply that some people pay more; markets have always produced unequal outcomes. It is that the mechanism becomes unknowable. In a market of posted prices, a high price is public information that competitors can undercut and shoppers can refuse. In a market of personalised prices, it is a private transaction between you and a model, invisible to everyone else, including the regulators, journalists and rival retailers who might otherwise discipline it. The discipline of the market depends on the price being a shared fact. Surveillance pricing dissolves the shared fact, and with it the discipline.
The Fairness Nobody Consented To
Set aside the question of whether you pay more or less. Ask instead the question the practice never lets you ask: what, exactly, is being used to decide. This is where personalised pricing stops being a story about money and becomes one about discrimination in the older and graver sense of the word.
A price built from inferred willingness to pay is a price built from a model of who you are, and the characteristics that feed such a model are not chosen for their moral acceptability. They are chosen because they predict. If income predicts willingness to pay, the model uses income, and if it can infer income from your postcode, your device, your browsing and the brands you buy, then it is charging you according to your wealth without ever asking your salary. If household size predicts inelastic demand, the model uses household size, which means a larger family, often a poorer one, may face systematically higher prices on essentials it cannot do without. The Instacart patents named age, gender, household size and income directly. Some are characteristics anti-discrimination law has spent a century learning to treat as illegitimate grounds for differential treatment. None is one an ordinary shopper would knowingly hand over as a reason to be charged more for milk.
The trouble is that the shopper never gets to decide. The whole design of surveillance pricing is that the grounds of differentiation are hidden. You cannot object to being priced on your gender if you do not know your gender is in the model. You cannot contest a markup based on the inference that you are housebound if you never learn the inference was made. The ordinary apparatus of fairness, the ability to know the reason for a decision and to challenge it, simply does not engage, because the reason is buried in a proprietary system and the decision arrives disguised as a fact of nature. A price, to the shopper, looks like something the world has handed down. It does not look like an accusation, a profile or a bet. But that is what, increasingly, it is.
This is the argument that Veena Dubal, professor of law at the University of California, Irvine, has developed across both the consumer and the labour sides of the same phenomenon. Writing in Governing magazine in April 2026, Dubal set out why AI should not be setting prices or wages, and why states needed to push back. The techniques now spreading through consumer pricing were pioneered on workers, in the ride-hailing and food-delivery platforms, where her earlier research documented what she named algorithmic wage discrimination: the practice of paying different workers different amounts for substantially the same work, with the wage personalised in real time according to dozens of behavioural signals invisible to the worker. The platform companies, she has observed, have been at the cutting edge of experimenting with ways to control people without it being obvious, and when those experiments work, they leach into other industries. Surveillance pricing is the consumer-facing twin of surveillance pay. Both rest on the same engine of behavioural inference. Both produce outcomes the affected person cannot predict, cannot explain and cannot contest.
The Revenue Nobody Mentions
Dubal's Governing piece adds a dimension that rarely surfaces in the consumer-protection framing: the state's own balance sheet. When algorithmic systems reclassify what would once have been straightforward taxable wages into a shifting patchwork of bonuses and incentives, calibrated worker by worker, the effect is not only to make individual incomes unpredictable. It is to erode the tax base on which public insurance depends. Dubal cites Connecticut, estimating that the state stands to lose around 60 million dollars a year in unemployment-insurance contributions as wages are restructured into forms that escape the payroll levy. The same opacity that lets a company extract a few extra pennies from a vulnerable shopper lets it shrink its obligations to the commons, and because the mechanism is granular and individualised, it is fiendishly hard for any tax authority to see, let alone challenge.
This is the quiet scandal beneath the loud one. The visible harm of surveillance pricing is the markup on your groceries. The invisible harm is what the same techniques do to the institutions that depend on legible, shared economic facts: tax systems, labour statistics, consumer-price indices, the apparatus by which a society measures and governs its own economy. An economy of personalised prices and wages becomes progressively harder to measure, because measurement assumes that prices and wages are public things. The spread of these tools from the gig platforms into healthcare, retail, logistics and customer service threatens not only individual fairness but the informational foundations of governance itself. An audit of 500 AI vendors her research points to found at least 20 at high risk of enabling surveillance-based pay, most already wired directly into employers' payroll and HR systems. The leach is well underway.
A Legal Framework That Is Mostly a Vacuum
If this sounds like the sort of thing the law would surely prohibit, the uncomfortable answer is that, for the most part, it does not. In April 2026, the legal-analysis service JD Supra carried a clear-eyed assessment, written by attorneys at the firm Holland and Knight. Their conclusion was blunt: there is no comprehensive federal statutory framework in the United States governing surveillance pricing. What exists instead is improvisation, the stretching of older authorities to cover a practice their drafters never imagined. Enforcement, where it happens at all, leans on Section 5 of the FTC Act, which prohibits unfair or deceptive practices, and on the agency's rule against unfair or deceptive fees. The authors noted that pricing-enforcement risk was no longer theoretical but an active priority. Yet active priority is not clear law. Section 5 was written to police deception in the abstract, not to answer whether a retailer may infer your income from your shopping habits and charge you accordingly, and the absence of a statute on that question leaves enormous room for argument, delay and retreat.
The patchwork that fills the federal vacuum is uneven and young, but filling fast. New York has moved fastest on disclosure, with a law requiring that when a price has been personalised using a consumer's data, the shopper must be told, in some versions with a stark warning that an algorithm set the price. The disclosure approach restores a measure of the visibility that surveillance pricing destroys, but it does not prohibit the practice, and a warning that everyone learns to ignore is a thin protection. The decisive shift in 2026 has been towards bans. By the spring, state lawmakers had introduced more than forty bills across at least twenty-four states to regulate personalised algorithmic pricing, outpacing the whole of 2025.
Maryland was the first to enact one, its governor, Wes Moore, signing the Protection From Predatory Pricing Act on 28 April 2026, effective 1 October. The statute, the first of its kind in the food sector, prohibits large food retailers and third-party delivery services from using personal data to set higher prices for particular consumers or classes. It carries penalties of up to 10,000 dollars per violation, rising to 25,000 dollars for repeat offenders, enforced by the state attorney general. It also carves out the practices the industry was most anxious to protect: loyalty schemes, voluntary membership discounts, genuine promotional offers and price differences attributable to objective costs such as shipping or tax.
Connecticut followed within weeks. Its bill, SB 4, passed the legislature on 4 May 2026, 141 to 6 in the House and 31 to 4 in the Senate, and Governor Ned Lamont signed it on 27 May as Public Act 26-64. Where Maryland's law reaches only the food sector, Connecticut's prohibits surveillance pricing, defined as setting a customised price for a consumer or group of consumers on the basis of personal data gathered through any technology, across retail generally and binding third-party delivery services. The act goes further still, establishing a state data-broker registry, a one-request mechanism for wiping a consumer's records across the whole industry, and a ban on the sale of precise geolocation data. Its provisions take effect on 1 October 2026, the same day as Maryland's.
The same season showed how easily such laws fail to arrive. Colorado's legislature passed the most ambitious measure of all, HB26-1210, which would have banned individualised price and wage setting based on surveillance data across every industry, on 8 May 2026. Governor Jared Polis vetoed it. His objection was not to the principle but to the reach: the bill, he wrote, took too broad an approach, capturing any technology that incidentally influences a price or wage amount rather than targeting unethical conduct, and would punish lower prices as readily as higher ones. The veto is an instructive counterpoint to the bills that passed: the friction these tools provoke reaches all the way to the governor's desk. California, meanwhile, kept moving: its surveillance-pricing bill cleared a key vote on 15 May 2026 and is still in progress.
Governing magazine's April 2026 reporting, and Dubal's argument within it, treated these state moves as the leading edge of a necessary legislative response rather than a settled solution. The pattern is familiar from the early history of data protection and of antitrust in the digital economy. The technology arrives at national, indeed global, scale. The law responds at the level of individual states, slowly, unevenly and with vigorous lobbying against every clause. The result, for now, is a map in which the legality of charging you a personalised price for a tin of beans depends substantially on which state line you happen to be standing behind, and in most of the country the answer remains that the practice is lawful, undisclosed and unmeasured.
What Europe Did Differently, and Did Not
Across the Atlantic, the legal starting point is different, though it is a mistake to imagine it amounts to a clean prohibition. The European Union confronted personalised pricing earlier and built a disclosure obligation into its consumer law through the Omnibus Directive of 2019, which took effect across member states in 2022. Under it, a trader must inform a consumer whenever the price they are being shown has been personalised on the basis of automated decision-making and profiling. The obligation is narrower than it sounds. It requires the seller to say that the price is personalised; it does not forbid the personalisation, and it does not require the seller to reveal what data went into it or how. A consumer told that a price has been tailored to them learns that they are being profiled without learning anything about the profile.
The heavier weapon in the European arsenal is data-protection law, and here the picture is genuinely contested. Article 22 of the General Data Protection Regulation gives individuals a right not to be subject to decisions based solely on automated processing that produce legal or similarly significant effects on them. Whether a personalised price counts as such a decision, and whether the regulation can therefore be read to require explicit consent before a shopper is priced by algorithm, is a question on which European lawyers have argued for years without settling. Some scholars contend that the GDPR, read seriously, enshrines something like a right to an impersonal price, a right to be quoted the same figure as everyone else unless you have genuinely agreed otherwise. Others regard that reading as aspirational. What is not in dispute is that European anti-discrimination law forbids using certain protected characteristics, of the kind that pricing models are perfectly capable of inferring, as the basis for differential treatment. The European framework, in other words, contains stronger raw materials than the American one, but it has not yet been assembled into a coherent answer to the specific harm. The United Kingdom, having left the EU before the Omnibus Directive bound it, is under no obligation to mirror even the disclosure rule, and its Competition and Markets Authority has approached the question through its broader work on online choice architecture and the manipulative design of digital interfaces rather than through a dedicated pricing statute.
The comparison yields a sober conclusion. No major jurisdiction has yet produced a settled, comprehensive answer to the question of when, if ever, a company may charge you a price calculated from a secret model of who you are. Europe has more tools and more disclosure. America has more enforcement appetite in some states and almost nothing in most. Everywhere, the technology is ahead of the law, and everywhere the burden of that gap falls on the individual shopper, who has neither the information to know what is happening nor the standing to do much when they find out.
The Asymmetry at the Heart of It
Strip the subject to its bones and what remains is an asymmetry of knowledge so steep that it makes a mockery of the idea of a transaction between equals. The seller knows the cost of the good, the price it shows you, the price it shows others, the model that produced your figure and the data that fed the model. You know the price it shows you. That is all. You cannot see the distribution of prices around you, the inputs, or the inference. You cannot even reliably tell whether personalisation is happening at all, because a personalised price and a non-personalised one look identical: both are just numbers on a screen. The market, classically conceived, was supposed to be an information system, aggregating dispersed knowledge into a public signal that coordinated the behaviour of strangers. Surveillance pricing inverts it. It turns the price from a signal the market sends to you into a signal the seller extracts from you, and does so silently, so that you go on reading the number as though it still carried its old public meaning.
This is why disclosure remedies, useful as they are, feel inadequate to the scale of the thing. Telling a shopper that their price has been personalised restores a sliver of the lost information, but it leaves the deeper asymmetry intact. It is rather like being told that a stranger has formed an opinion of your character without being told what the opinion is or what evidence it rests on. The grievance is not merely that the price was tailored. It is that it was tailored using a portrait of you that you did not sit for, that you cannot see and that may be wrong, unfair or built from characteristics you would never have agreed to be judged by. The ordinary person's intuition that there is something improper here is not naivety about how markets work. It is an accurate perception that a hard-won feature of how markets are supposed to work, the shared and public price, is being quietly removed, with nothing put in its place to protect them.
Switching the Lights Back On
What, then, is the ordinary person at the invisible checkout to do? Honesty requires admitting that individual self-defence is mostly futile. Clearing cookies, browsing privately, comparing prices across devices: these are the folk remedies of a simpler era of price discrimination, and against a system that fuses dozens of inferred signals they offer little. The man who deleted his cookies on Amazon in 2000 found a cheaper DVD because the discrimination then was crude. It is not crude now, and the burden of evading it cannot reasonably be placed on the shopper. A person should not have to conduct counter-surveillance against their grocer to be charged a fair price for bread.
The more honest answer is that this is a collective problem requiring collective tools, and the encouraging part of the story is that those tools are beginning, haltingly, to appear. The Instacart episode is the clearest demonstration of the mechanism that actually works, which is exposure. The company did not stop because the law compelled it. There was, in the relevant sense, no law to compel it. It stopped because an investigation made the practice visible, and visibility was intolerable to a business that depended on shoppers believing the price was the price. Lindsay Owens of the Groundwork Collaborative put the dynamic with precision when she said that once the curtain was pulled back, the company had no choice but to close the lab. Surveillance pricing is a practice that cannot survive being seen. That is its great vulnerability, and it points directly at the remedy.
The remedy has three reinforcing layers. The first is sunlight, the dogged work of investigators, researchers and regulators in dragging an invisible practice into view, because each exposure raises the reputational cost of doing it. The FTC study, the Consumer Reports investigation and the work of scholars like Dubal are instances of the same act: making the hidden price visible so it can be argued about. The second is disclosure as a legal default, the New York and European approach of requiring sellers to declare when a price has been personalised, imperfect but better than silence. The third, on which the rest depend, is substantive law of the kind Maryland and Connecticut have now enacted: rules that do not merely require disclosure but forbid the use of certain data and inferences to set the price of essentials, and give a public enforcer the teeth to make the prohibition real. Colorado's veto shows that this third layer is the hardest to lay, the one over which the fight is fiercest.
None of this will arrive quickly or cleanly, and the lobbying against every line of it will be intense, because the prize for the seller is enormous and the constituency for the shopper is diffuse. But the direction of travel is set by a simple fact that no amount of optimisation can engineer away. People do not want to be charged according to a secret estimate of how much they can be made to bear. The Consumer Reports survey found the objection close to unanimous, and it cut across every reason a company might offer. That near-universal refusal is the political bedrock on which any durable response will be built. The invisible price depends, in the end, on staying invisible. The work of the coming years, in legislatures and regulators and newsrooms alike, is to ensure that it cannot.
The next time you tap to confirm an order and the total looks about right, hold for a second the thought that you cannot verify it is right, because right has quietly stopped meaning the same thing for everyone. That second of doubt is not paranoia. It is the appropriate response of a citizen to a market that has learned to read them and has not asked permission. The price you see may be the price everyone sees. It may not. That you can no longer tell is the whole problem, and reclaiming the ability to tell is the whole of the answer.
References
- Consumer Reports. “Instacart's AI-Enabled Pricing Experiments May Be Inflating Your Grocery Bill, CR and Groundwork Collaborative Investigation Finds.” December 2025. https://www.consumerreports.org/money/questionable-business-practices/instacart-ai-pricing-experiment-inflating-grocery-bills-a1142182490/
- Consumer Reports. “New Report Exposes Instacart's Hidden Price Games.” Press release. December 2025. https://www.consumerreports.org/media-room/press-releases/2025/12/new-report-exposes-instacarts-hidden-price-games/
- Consumer Reports. “Instacart Stops AI Pricing Tests.” December 2025. https://www.consumerreports.org/money/questionable-business-practices/instacart-stops-ai-pricing-experiments-a1176475852/
- CBS News. “Instacart to end AI price tests for retailers following investigation.” 23 December 2025. https://www.cbsnews.com/news/instacart-ends-ai-price-testing-tool-eversight/
- Grocery Dive. “Instacart ends controversial price tests.” December 2025. https://www.grocerydive.com/news/instacart-ends-controversial-price-tests/808490/
- Veena Dubal. “AI Shouldn't Be Setting Prices or Wages. States Need to Push Back.” Governing. 20 April 2026. https://www.governing.com/workforce/ai-shouldnt-be-setting-prices-or-wages-states-need-to-push-back
- Veena Dubal. “On Algorithmic Wage Discrimination.” Columbia Law Review 123(7), 2023. https://columbialawreview.org/content/on-algorithmic-wage-discrimination/
- Christopher Armstrong, Benjamin Genn, Ashley Joyner Chavous and Kwamina Thomas Williford (Holland and Knight LLP). “Surveillance Pricing, AI Pricing Tools and the Push for Price Transparency.” JD Supra. 28 April 2026. https://www.jdsupra.com/legalnews/surveillance-pricing-ai-pricing-tools-8489430/
- Federal Trade Commission. “FTC Surveillance Pricing Study Indicates Wide Range of Personal Data Used to Set Individualized Consumer Prices.” January 2025. https://www.ftc.gov/news-events/news/press-releases/2025/01/ftc-surveillance-pricing-study-indicates-wide-range-personal-data-used-set-individualized-consumer
- Federal Trade Commission. “Issue Spotlight: The Rise of Surveillance Pricing.” January 2025. https://www.ftc.gov/system/files/ftc_gov/pdf/sp6b-issue-spotlight.pdf
- Office of Governor Wes Moore. “Governor Moore Signs Legislation to Protect Marylanders' Pocketbooks in Grocery Stores.” 28 April 2026. https://governor.maryland.gov/news/press-releases/governor-moore-signs-legislation-protect-marylanders-pocketbooks-grocery-stores-safeguard-voting
- Skadden, Arps, Slate, Meagher and Flom LLP. “Maryland Becomes the First State to Restrict Surveillance Pricing in the Food Industry.” May 2026. https://www.skadden.com/insights/publications/2026/05/maryland-becomes-the-first-state-to-restrict-surveillance-pricing
- Greenberg Traurig LLP. “Maryland Enacts Food-Sector Personalized Pricing Law.” May 2026. https://www.gtlaw.com/en/insights/2026/5/maryland-enacts-food-sector-personalized-pricing-law
- Future of Privacy Forum. “Third Time's the Charm: Connecticut Enacts Annual Privacy Update.” May 2026. https://fpf.org/blog/third-times-the-charm-connecticut-enacts-annual-privacy-update/
- Covington and Burling LLP. “Connecticut Enacts Omnibus Privacy Law.” Inside Privacy. May 2026. https://www.insideprivacy.com/state-privacy/connecticut-enacts-omnibus-privacy-law/
- Colorado Newsline. “Colorado bill to ban surveillance prices, wages vetoed by Gov. Polis.” May 2026. https://coloradonewsline.com/briefs/surveillance-pricing-bill-vetoed/
- CalMatters. “Why surveillance pricing bans are suddenly gaining traction this year (and not just in California).” May 2026. https://calmatters.org/economy/technology/2026/05/why-surveillance-pricing-bans-are-suddenly-gaining-traction-this-year-and-not-just-in-california/
- Covington and Burling LLP. “State Lawmakers Introduce New Wave of Personalized Algorithmic Pricing Bills.” Inside Privacy. 2026. https://www.insideprivacy.com/artificial-intelligence/state-lawmakers-introduce-new-wave-of-personalized-algorithmic-pricing-bills/
- European Parliament. “Personalised pricing.” Study, IPOL, 2022. https://www.europarl.europa.eu/RegData/etudes/STUD/2022/734008/IPOL_STU(2022)734008_EN.pdf
- BEUC (The European Consumer Organisation). “Price personalisation.” 2023. https://www.beuc.eu/sites/default/files/publications/BEUC-X-2023-097_Price_personalisation.pdf
- Frederik Zuiderveen Borgesius and others. “The GDPR enshrines the right to the impersonal price.” Computer Law and Security Review, 2022. https://www.sciencedirect.com/science/article/pii/S0267364922000085
- Washington Center for Equitable Growth. “How artificial intelligence uncouples hard work from fair wages through 'surveillance pay' practices, and how to fix it.” 2026. https://equitablegrowth.org/how-artificial-intelligence-uncouples-hard-work-from-fair-wages-through-surveillance-pay-practices-and-how-to-fix-it/
- OECD. “Personalised Pricing in the Digital Era.” DAF/COMP(2018)13. 2018. https://one.oecd.org/document/DAF/COMP/WD(2018)150/en/pdf
Tim Green
UK-based Systems Theorist & Independent Technology Writer
Tim explores the intersections of artificial intelligence, decentralised cognition, and posthuman ethics. His work, published at smarterarticles.co.uk, challenges dominant narratives of technological progress while proposing interdisciplinary frameworks for collective intelligence and digital stewardship.
His writing has been featured on Ground News and shared by independent researchers across both academic and technological communities.
ORCID: 0009-0002-0156-9795
Email: tim@smarterarticles.co.uk
Listen to the free weekly SmarterArticles Podcast
