CHE | Getty
In a landmark copyright settlement — believed to be the largest of its kind in history — the AI company Anthropic recently agreed to pay approximately $1.5 billion to resolve claims that it illegally trained its chatbot Claude on roughly 500,000 books without first seeking permission from their authors or publishers. Call it progress; call it a down payment; but please don’t call it a solution. It’s not. While the settlement addresses piracy from shadow libraries, it does nothing to deter the scraping of peer-reviewed journals, open-access books, working papers, and disciplinary literatures that actually teach the most powerful AI systems how to reason.