Nobody sets out to build a manipulative interface. These patterns arrive as growth experiments nobody wants to argue with, defended one meeting at a time by people doing their jobs well. In September 2025, the bill came due for the largest of them.
The habits are not accidents
Anyone who has worked in this field long enough has shipped one of these. We have. That is not a confession dressed up as humility, it is the actual point, because the interesting question was never why bad people build bad interfaces. It is why good ones do.
Here is how it usually happens. Someone brings a number to a meeting. The number is up. The change that moved it was small: a heavier button, a quieter link, one more screen between a customer and the door. Nobody wants to be the person who argues with a number, and honestly, the argument is hard to make. You are asking a team to trade something measurable today for something unmeasurable later.
So the change ships. Then another one. Every step defensible on its own, and the whole path only visible from a distance.
That is how trust erodes. Not in a decision. In a hundred small defenses, each one reasonable.
Worth noting what you will not find anywhere in the public record: a named designer at a named company, on the record, saying they meant to do it. That admission does not exist. Not because everyone is lying, but because almost nobody experienced it as a decision.
What the field actually calls this
The vocabulary came from one person. Harry Brignull, a UX researcher with a PhD in cognitive science, coined the term “dark patterns” in 2010 to name what he described as “manipulative, deceptive and coercive design techniques.” He has since published a book on them, and the taxonomy he built moved out of the design community and into law.
Nielsen Norman Group put a working definition on it. Their researcher Maria Rosala defines a deceptive pattern as one that will “prompt users to take an action that benefits the company” at the user’s expense.
Hold onto that phrasing. Not tricks the user. Benefits the company. That is the part that makes these things survive review, because on any given Tuesday, benefiting the company is the job.
The bill came due in September 2025
The Federal Trade Commission named the category in a 2022 staff report. For three years the argument stayed academic. Then it stopped.
Amazon settled with the FTC for $2.5 billion over how Prime enrollment and cancellation were designed. A billion of that was a civil penalty, the largest ever in a case involving an FTC rule violation. The remaining $1.5 billion went back to roughly 35 million people who had signed up for something they did not mean to buy. The settlement landed three days into trial.
“putting billions of dollars back into Americans’ pockets”
Andrew N. Ferguson, Chairman, U.S. Federal Trade Commission, September 25, 2025
Read that as a design review with subpoena power.
Five habits, and the defense each one comes with
None of these is exotic. That is the problem with them.
-
01. The decline button written as an insult
The industry calls it confirmshaming. “No thanks, I don’t want to save money.” The defense: it lifts opt-in, and it does. The cost: the Amazon order now specifically bars this construction. A button worded to make declining feel foolish stopped being a clever nudge somewhere around last September. It became a compliance exhibit.
-
02. The exit that costs more than the entrance
One click to join. Six screens, two save-offers, and a phone call to leave. The defense: friction saves accounts, and for a quarter or two it does. The cost: this asymmetry was the spine of the government’s case. A federal judge ruled in September 2025 that collecting billing details before clearly disclosing terms violated the Restore Online Shoppers’ Confidence Act.
-
03. The default that is not neutral
Pre-checked boxes. Bundled consent. Terms behind a link nobody opens. The defense: anyone can uncheck it. The cost: express informed consent became a legal standard rather than a design preference, and a default is a decision you made on someone’s behalf while telling yourself they made it.
-
04. The account wall standing in front of the value
Register first, see the thing you came for second. The defense: the data funds a better experience down the line, which is sometimes even true. The cost: the room you are making that argument in. Pew Research found 81% of American adults are concerned about how companies use their data, 73% feel they have little or no control over it, and 67% say they understand little to nothing about what happens to it, up from 59% four years earlier.
-
05. Urgency that isn’t real
The countdown that resets on refresh. The “three people are viewing this” on an empty page. The defense: it converts, and it does, once. The cost: everything you say afterward. A customer who catches one invented number has no reliable way to sort your true claims from your decorative ones, so they stop trying.
“You can design your way into a conversion. You can’t design your way out of being distrusted.”
David Keyes, Founder & CEO, HAUS XXIV
The rule fell. The exposure didn’t.
There is a wrinkle worth getting right, because a lot of teams read the 2025 headlines and heard permission.
The FTC finalized its expanded Negative Option Rule in October 2024, the one everyone called Click to Cancel. In July 2025 the Eighth Circuit vacated it in full. Not on the merits of the idea, on procedure, because the agency had skipped a required economic analysis for a rule of that weight.
That looked like a reprieve. It wasn’t one.
ROSCA is still good law and the FTC kept enforcing under it, which is exactly how the Amazon judgment landed two months after the rule was struck down. The agency brought a subscription case against Uber the same year. And in March 2026 the FTC opened new rulemaking to rebuild the rule properly, with public comment closing that April.
The short version for anyone running a subscription business: the referee stepped off the field for about eight weeks, and the penalties got bigger, not smaller.
A test that takes four seconds
We use a simple one internally, and it is worth stealing.
Put your signup and your cancellation on the same screen and count the steps. That’s it. The gap between those two numbers is the most honest metric in the building, and nobody has to run a study to read it.
It works because it is not really a UX question. It is the seventh principle in our Code of 24™, conviction over convenience, pointed at an interface. We don’t perform our values, we practice them, and an interface is where a company’s values become legible to a stranger in about four seconds. Discipline eighteen puts it plainly: UX and UI engineered for real humans.
None of the five habits above came from cynicism. They came from pressure, and from the fact that the case against them is slow while the case for them fits on a slide. That is what makes refusing them a discipline rather than a preference. The harder choice happens on a Tuesday, in a backlog grooming session, when the easier one is right there and nobody would have noticed either way.
Safe doesn’t survive here.
