Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

False Positive Paradox

Also Known As: Base rate neglect in testing Screening paradox
Aspect ID: false_positive_paradox

Definition

The false positive paradox occurs when a highly accurate test applied to a rare condition produces more false positives than true positives in absolute terms. Even a test with 99% sensitivity and 99% specificity will produce one false positive for every true positive when testing a population with 1% prevalence, and ten false positives for every true positive at 0.1% prevalence.

Examples

A disease affects 1 in 1,000 people. A test has 99% sensitivity and 99% specificity. Testing 100,000 people: 100 true cases, of which 99 test positive. But 99,900 healthy people test, of which 999 test positive (false positives). There are 10 false positives for every true positive.

An airport security algorithm flags potential threats with 99% accuracy and a false positive rate of just 1%. On a day with 10,000 travelers, if only 10 are genuine threats, the system correctly catches 9 of them — but also wrongly detains 100 innocent passengers. For every real threat identified, roughly 11 innocent people are flagged alongside them.

A social media platform deploys an AI to detect bot accounts, claiming 98% accuracy. If only 0.5% of its 10 million users are bots — that's 50,000 bots — the system correctly identifies 49,000 of them but also falsely flags 199,000 real users. The overwhelming majority of accounts banned are actually legitimate human users.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Is the test being applied to a low-prevalence condition?

    Type: binary
  2. 2

    Is the specificity of the test high enough to prevent false positives from dominating true positives?

    Type: binary
  3. 3

    Is the positive predictive value (PPV) calculated using the actual population prevalence?

    Type: binary
  4. 4

    Are absolute counts of true positives versus false positives reported, not just sensitivity and specificity?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.