How Statistical Watermarks Detect AI-Generated Text

Computerphile · Explainer · 14 min · 2023 · YouTube title: Ch(e)at GPT? - Computerphile

Why it's here

Linked as the approachable explanation whenever HN discusses detecting or watermarking AI-generated text.

What it covers

Computerphile video in which Mike Pound explains a University of Maryland paper on watermarking LLM output. The method biases token choices so generated text can later be detected statistically.