What Are Dark Patterns and What Types Does Dark Patterns Hall of Shame Cover?

Updated

Dark patterns are deceptive or manipulative user-interface designs that push people into choices they would not otherwise make—such as subscribing, sharing data, or staying in a service they want to leave. The Dark Patterns Hall of Shame is a case library that documents these designs in real products, tagging each entry with a pattern type and explaining how the manipulation works. It is useful if you want to recognize these tactics in your own accounts or cite concrete examples; it is not a legal authority or a complete taxonomy of every pattern in academic research.

How the site organizes its cases

Each entry pairs a named pattern with a specific company and a short write-up. The pattern label tells you the mechanism; the case tells you how it appeared in practice. From the site's own coverage, the recurring types include:

Pattern What it does Example on the site
Preselection Options are pre-checked so inaction equals consent Phosus pre-selecting marketing emails; Groww pre-checking a credit-score permission box
Privacy Zuckering Design nudges you into sharing more data than intended LinkedIn training AI on user content
Obstruction Steps are added to make a desired action hard Amazon's Audible cancellation flow
Roach Motel Easy to enter, hard to leave LinkedIn Premium cancellation
Nagging Repeated prompts to do something you already declined Reddit pushing its app; TikTok requesting notification access
Trick Questions Wording or layout that misleads your answer Mailchimp's sign-up process
Cookie Wall Access is blocked until you accept cookies HRAcuity requiring cookie consent to view a video
Fake Scarcity Artificial urgency or limited-supply pressure StubHub ticket sales (per a D.C. attorney general lawsuit)
Misdirection Attention is drawn away from the consequential choice Listed alongside obstruction in the Audible case

What the individual cases show

Preselection appears most often in the library. The pattern works because the default is treated as consent: a pre-ticked box, an auto-enabled setting, or an opt-out buried after the fact. Figma's 2025 AI rollout is described as automatically opting all users in to have their design files used for AI training—so the "choice" existed but the default did the deciding.

Obstruction and Roach Motel cover the exit side. The Audible cancellation write-up notes the flow follows classic patterns "designed to complicate the process," and the LinkedIn Premium piece frames cancellation as a case where a design promising ease serves a different purpose. The practical takeaway: when leaving is harder than joining, expect extra steps, retention offers, or hidden paths.

Nagging is the mildest but most frequent. Reddit's repeated app-download prompts and TikTok's repeated notification requests are both framed as pushing past a clear decline.

Cookie Wall and Trick Questions sit closer to consent and comprehension. A cookie wall trades access for consent; a trick question makes the "right" answer ambiguous through wording or layout.

How to use it

  1. Identify the mechanism first. Before reading the case, ask what the interface is doing to your default behavior—pre-ticking, hiding, repeating, or blocking.
  2. Check the pattern tag. The site's labels (Preselection, Nagging, Obstruction, etc.) give you vocabulary to describe what you're seeing in other products.
  3. Read the case for the specific flow. The value is in the concrete steps—where the opt-out lives, what the cancellation path looks like—not the label alone.
  4. Verify before acting on a claim. Entries are dated and some reference external events (like the StubHub lawsuit). For anything involving your account or a legal matter, confirm against the company's current settings or official sources.

Limits to keep in mind

The library is a curated collection of examples, not an exhaustive or peer-reviewed classification. Coverage is weighted toward well-known tech products and English-language services, and entries reflect the state of a product at the time of writing—interfaces change. Treat it as a starting point for recognizing manipulation, and check current behavior yourself before drawing conclusions about any specific service.