CNN sues Perplexity AI over mass copyright scraping of 17,000 stories

CNN’s suit against Perplexity is a shot across the bow for any AI company whose value proposition depends on repackaging premium news. What makes this case different is not just the volume of allegedly scraped content, but the fact that CNN is arguing Perplexity’s outputs can be near‑verbatim reproductions that keep users off CNN’s own properties. If courts accept that framing, it could force a re‑architecture of how retrieval‑augmented generation and AI search engines handle paywalled or licensed material.

In the race to AGI, this matters because frontier labs and search‑style products depend on wide‑open access to the public web and, implicitly, to journalism. A legal environment in which high‑quality corpora can only be accessed via expensive, tightly‑scoped licenses would raise training and serving costs, advantage better‑capitalized incumbents, and potentially slow independent entrants. It also accelerates a broader shift: publishers realizing that their archives are scarce strategic assets in an AI economy, not just back catalogues for human readers.