AI Race Shifts to Data: Old Books Become New Training Goldmine

Olivier__OG · x · 2026-07-31

As public internet data becomes polluted with AI-generated content, the AI race is shifting from models to scarce human data. Pre-internet books are suddenly valuable for being clean training material.

Reports on Anthropic's book-scanning project reveal how physical books are sourced and sometimes destroyed to create private datasets. Owning unique, high-quality data is becoming the real moat, suggesting future AI advantage belongs to companies with the cleanest, rarest data.

Original post →

More from AGI Musings

AGI Musings channel →