Policy & Ethicsllmtraining dataopenaicopyright
Britannica Sues OpenAI Over Copyrighted Content
9.2
Encyclopaedia Britannica and Merriam-Webster filed a lawsuit against OpenAI on Friday, alleging the company used their copyrighted content to train GPT-4 and produced near-verbatim outputs. The publishers submitted side-by-side examples and say OpenAI’s responses cannibalize their web traffic by substituting traditional referrals. The suit follows similar cases by The New York Times and a $1.5 billion Anthropic settlement.
Key Points
- 1Alleges OpenAI trained models on Britannica and Merriam-Webster copyrighted text, producing near-verbatim outputs.
- 2Claims such copying reduces Britannica’s web traffic and substitutes search referrals, threatening publisher revenue.
- 3Raises legal precedent risk for AI firms; practitioners must audit training data and licensing practices.
Scoring Rationale
High-impact legal challenge to LLM training practices; strong official filings but broader industry outcomes remain uncertain.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
High-Value Overnight OrdersEasyDelivered International ShipmentsMediumOn-Time Delivery Rate by CarrierHard
250 free problems · No credit card
See all Logistics & Shipping problems
