According to Press Gazette (July 16, 2026), Google is scrubbing thousands of legitimate news articles from search results after receiving fraudulent DMCA takedown notices — a mechanism built to police piracy that is instead being used, per the report, to bury inconvenient journalism. The mechanics are familiar to anyone tracking the data economy: the DMCA’s notice-and-takedown regime relies on platforms acting fast and asking questions later, which means a bad-faith claimant can weaponize copyright law against a rival, a critic, or a story they simply don’t like, with little upfront cost.
For the data and AI-licensing world, this is more than a press-freedom footnote. The same takedown plumbing that erases search results is the plumbing publishers and platforms increasingly lean on to police unauthorized scraping and AI training use. If spurious notices can already strip real reporting from Google at scale, the incentive to game that system will only grow as copyright claims become currency in AI licensing disputes — whether to suppress competitors’ content, manipulate discoverability, or pressure outlets into deals.
A takedown system built to fight piracy is proving just as effective at fighting inconvenient facts.
Watch for whether Google tightens its DMCA review process or publishes more transparency data on rejected versus honored claims — and whether regulators or courts start treating bad-faith takedowns as their own form of actionable abuse, distinct from the underlying copyright fights over AI training data.
Google is removing thousands of articles from search due to spurious DMCA requests.
