Agentic data harvesting on prices will also allow the kind of "corner the market" manipulation that skyrocketed dram memory prices.

techradar.com/pro/dram-prices-…

forbes.com/sites/timbajarin/20…

cnbc.com/2026/01/10/micron-ai-…

medium.com/@sahendra_/ram-cris…

tomshardware.com/pc-components…

A billionaire or hedge fund can buy year's worth of critical minerals or industrial inputs and trigger global economic warfare.

AI memory is sold out, causing an unprecedented surge in prices

Three primary memory vendors — Micron, SK Hynix and Samsung Electronics — make up nearly the entire RAM market, and they're benefitting from this shortage.

^{Kif Leswing (CNBC)}

Some of that is merely because websites have inconsistent "semantics" - literally things like the "buy" button being called something other than "buy button" in the HTML code. But there's a far more profound problem with agentic shopping, which is that companies *deliberately obfuscate their prices*.

This is how junk fees work, and why they're so destructive. Say you're a hotel providing your rate-card to an online travel website.

11/

You know that travelers are going to search for hotels by city and amenities, and then sort the resulting list by price. If you hide your final price - by surprising the user with a bunch of junk fees at checkout, or, better yet, *after* they arrive and put their credit-card down at reception - you are going to be at the top of that list. Your hotel will seem like the cheapest, best option.

12/

But of course, it's not. From Ticketmaster to car rentals, hotels to discount airlines, rental apartments to cellular plans, the *real* price is withheld until the very last instant, whereupon it shoots up to levels that are absolutely uncompetitive. But because these companies are able to engage in deceptive advertising, they *look* cheaper.

13/

And of course, crooked offers drive out honest ones. The honest hotel that provides a *true* rate card, reflecting the all-in price, ends up at the bottom of the price-sorted list, rents no rooms, and goes out of business (or pivots to lying about its prices, too).

Online sellers *do not want* to expose their true prices to comparison shopping services. They benefit from lying to those services.

14/

For *decades*, technologists dreamed of building a "semantic web" where everyone exposes true and accurate machine-readable manifests of their content to facilitate indexing, search and data-mining:

people.well.com/user/doctorow/…

This has failed. It's failed because lying is often more profitable than telling the truth, and because lying to computers is easier than lying to people, and because once a market is dominated by liars, everyone has to lie, or be pushed out of the market.

15/

Of course, it would be *really cool* if everyone diligently marked up everything they put into the public sphere with accurate metadata. But there are lots of *really cool* things you could do if you could get everyone else to change how they do things and arrange their affairs to your convenience.

16/

Imagine how great it would be if you could just get everyone to board an airplane from back to front, or to stand right and walk left on escalators, or to put on headphones when using their phones in public.

Wanting it badly is not enough. People have lots of reasons for doing things in suboptimal ways. Often the reason is that it's suboptimal for you, but just peachy for *them*.

17/

Google says that it's going to get every website in the world to expose accurate rate cards to its chatbots to facilitate agentic AI. Google is also incapable of preventing "search engine optimization" companies from tricking it into showing bullshit at the top of the results for common queries:

pluralistic.net/2024/05/03/key…

18/

Google somehow thinks that the companies that spend millions of dollars trying to trick its crawler won't also spend millions of dollars trying to trick its chatbot - and they're providing the internet with a tool to inject lies straight into the chatbot's input hopper.

19/

But UCP isn't just a way for companies to tell Google what their prices are. As Stoller points out, UCP will also sell merchants the ability to have Gemini *set* prices on their products, using Google's surveillance data, through "dynamic pricing" (another euphemism for "surveillance pricing").

This decade has seen the rise and rise of price "clearinghouses" - companies that offer price "consulting" to direct competitors in a market.

20/

Nominally, this is just a case of two competitors shopping with the same supplier - like Procter and Gamble and Unilever buying their high-fructose corn-syrup from the same company.

But it's actually far more sinister. "Clearinghouses" like Realpage - a company that "advises" landlords on rental rates - allow all the major competitors in a market to collude to raise prices in lockstep.

21/

A Realpage landlord that ignores the service's "advice" and gives a tenant a break on the rent will be excluded from Realpage's service. The rental markets that Realpage dominates have seen major increases in rental rates:

pluralistic.net/2025/10/09/pri…

Google's "direct pricing" offering will allow all comers to have Google set their prices for them, based on Google's surveillance data. That includes direct competitors.

22/

As Stoller points out, both Nike and Reebok are Google advertisers. If they let Google price their sneakers, Google can raise prices across the market in lockstep.

Despite how much everyone hates this garbage, neoclassical economists and their apologists in the legal profession continue to insist that surveillance pricing is "efficient."

23/

Stoller points to a law review article called "Antitrust After the Coming Wave," written by antitrust law prof and Google lawyer Daniel Crane:

nyulawreview.org/issues/volume…

24/

Antitrust After the Coming Wave - NYU Law Review

A coming wave of general-purpose technologies, including artificial intelligence (“AI”), robotics, quantum computing, synthetic biology, energy expansion, and nanotechnology, is likely to fundamentally reshape the economy and erode the assumptions on…

^{Shawn Young (NYU Law Review)}

Crane argues that AI will kill antitrust law because AI favors monopolies, and argues "that we should forget about promoting competition or costs, and instead enact a new Soviet-style regime, one in which the government would merely direct a monopolist’s 'AI to maximize social welfare and allocate the surplus created among different stakeholders of the firm.'"

25/

This is a planned economy, but it's one in which the planning is done by monopolists who are - somehow, implausibly - so biddable that governments can delegate the power to decide what we can buy and sell, what we can afford and who can afford it, and rein them in if they get it wrong.

26/

In 1890, Senator John Sherman was stumping for the Sherman Act, America's first antitrust law. On the Senate floor, he declared:

> If we will not endure a King as a political power we should not endure a King over the production, transportation, and sale of the necessaries of life. If we would not submit to an emperor we should not submit to an autocrat of trade with power to prevent competition and to fix the price of any commodity.

pluralistic.net/2022/02/20/we-…

27/

Google thinks that it has finally found a profitable use for AI. It thinks that it will be the first company to *make* money on AI, by harnessing that AI to a market-rigging, price-gouging monopoly that turns Google's software into Sherman's "autocrat of trade."

28/