AntiPattern Logo
AntiPatternAI UX Lab

Pattern Catalog

Browse the most common AI UX failures by category, then study the corrected response pattern. Each entry provides a simulation, diagnosis, and checklist.

15 results15 total patterns6 categories

How to use the catalog

  1. 1. Pick a category that matches your UX risk.
  2. 2. Run the simulation to observe the failure.
  3. 3. Apply the checklist to design the fix.
New patterns are added as AI UX evolves.
Feedback & State
Anti-pattern

Infinite Typing Indicator

The system displays a typing or loading indicator indefinitely without receiving data.

Explore pattern
Trust & Transparency
Anti-pattern

Fake Confidence (Hallucination)

The AI presents incorrect or guarded information with absolute certainty.

Explore pattern
Errors & Refusals
Anti-pattern

Silent Failure

The system fails to perform an action but gives no visual indication of the error.

Explore pattern
Errors & Refusals
Anti-pattern

No Refusal State

The AI refuses a request without explanation or fallback.

Explore pattern
Communication & Tone
Anti-pattern

Over-Apologetic AI

The AI apologizes excessively for minor issues or even for user errors.

Explore pattern
Control & Agency
Anti-pattern

Irreversible Actions

The AI takes a destructive action (sending email, deleting file) without confirmation.

Explore pattern
Privacy & Consent
Anti-pattern

Hidden Data Usage

The system uses user data for training or context sharing without explicit consent.

Explore pattern
Feedback & State
Anti-pattern

Vague Loading States

The system implies it's 'working' but gives no indication of what it's actually doing.

Explore pattern
Control & Agency
Anti-pattern

One-Shot Interactions Only

The system forces a new conversation for every request, preventing refinement.

Explore pattern
Control & Agency
Anti-pattern

Uninterruptible Responses

The user cannot stop the AI while it is generating, forcing them to wait.

Explore pattern
Context & Memory
Anti-pattern

Context Amnesia

The AI fails to recall information provided in previous turns of the conversation.

Explore pattern
Onboarding
Anti-pattern

The Empty State Void

The interface provides no guidance, prompts, or capabilities at the start of a session.

Explore pattern
Control & Agency
Anti-pattern

Format Refusal (Lazy Output)

The AI ignores specific formatting instructions (JSON, Table, Markdown) and outputs conversational text.

Explore pattern
Feedback & State
Anti-pattern

Zombie Session

The UI allows interaction after the backend session has expired, leading to lost input.

Explore pattern
Security
Anti-pattern

Prompt Injection Vulnerability

The AI is easily manipulated into breaking its operating constraints or persona.

Explore pattern