Skip to content

Rival Research · Corpus Analysis

Ghosts in the Machine

What 2.14 million words from 250 AI models reveal about machine imagination. We asked every major model the same prompts and read every answer. The patterns in what they make up are stranger than the mistakes.
250 models·7,877 responses·2.14M words·March 2026

The most hallucinated name

Every AI has a favorite scientist. Her name is Chen.

Across unrelated prompts, with no name in the input, models kept inventing the same character. 279 times they reached for someone named Chen.

279

times models invented a character named "Chen". Nine model families produced "Sarah Chen" independently, with no prompt that mentioned the name.

No training set we know of explains it.

The Chen variants

Professor Chen
152
Sarah Chen
78
Prof. Chen
30
Alex Chen
19

Ask for a story, a sentience test, or an interview and the same names keep appearing. The collective imagination of 250 different models is smaller than one writer's.

Professor Chen
152
Aria
142
Sir Reginald
129
Sarah Chen
78
Professor Thorne
36
Professor Rivera
32
Professor Voss
31
Anna Reid
31
Elena Vasquez
27
Professor Vance
26

Convergence

Ask 250 models for a joke

You will mostly get one joke. 42% of models tell the exact same one, word for word.

42%

of 159 models tell the identical atoms joke: "Why don't scientists trust atoms? Because they make up everything."

Another 36% tell the scarecrow joke. AI draws from a tiny pool.

Which joke, share of models

Atoms joke
42%
Scarecrow joke
36%
Impasta joke
19%
Other
3%

The full report

60+ slides of where AI imagination breaks

This page is a taste. The full AI Hallucination Index goes deep on character hallucinations, per-provider fingerprints, the confident-but-false facts models assert, safety benchmarks, writing patterns, and cultural biases across all 250 models.

The full report

The AI Hallucination Index 2026

60+ data-driven slides on what 2.14M words from 250 models reveal about machine imagination, hallucination, and bias.

  • 2.14M words analyzed across 250 models
  • Per-model + per-provider hallucination fingerprints
  • Character hallucinations and confident-but-false facts
  • Safety benchmarks, writing patterns, and cultural biases
$960+ slide PDF
Launch price · over 50% offGet the full report

Secure checkout via Lemon Squeezy. Instant download.

Read the free sample (PDF)

Method

How we measured this

Every model answered the same standardized prompts on rival.tips. We stripped HTML and code, then counted, categorized, and cross-referenced every word: 2.14 million of them across 7,877 responses from 250 models.

Hallucinated characters are names that appear in outputs with no corresponding name in the prompt. Joke convergence is measured on the subset of models asked for a joke (159).

Sign in