Google suffered a court loss last week in its attempt to use the Digital Millennium Copyright Act to stop SerpApi, a company that scrapes search results and resells the data, from accessing its platform. Google has confirmed it will keep fighting, and Reddit, for reasons the lawsuit has not yet made plain, is fighting alongside it. SerpApi's response to its court win was short: Google and Reddit do not own the internet.
What Google is actually arguing
Google filed the complaint against SerpApi last December. The accusation is that SerpApi circumvented Google's anti-scraping technology and then monetized the results through what it markets as a Google Search API, a service Google describes as unauthorized.
The legal theory is worth examining closely. Google is not relying on a simple terms-of-service claim. It is reaching for the DMCA, a copyright statute typically used against tools that break digital locks on films or software. Google's argument is that its anti-scraping barriers protect copyrighted content in search results, and that bypassing them constitutes unlawful circumvention. Part of what makes the DMCA argument viable, Google contends, is that some content in its results is licensed from third parties. Knowledge panels, the information boxes that appear in results for well-known people or entities, can include such licensed material. SerpApi's activity, in Google's view, threatened those licensing relationships.
Reddit's position, and why it matters
The source notes Reddit is "curiously along for the ride" without spelling out the precise terms of its involvement. Both platforms have become targets for AI data collection at scale. The outcome here could shape how much control large platforms retain over content that AI systems train on or access in real time.
The counterargument SerpApi is currently winning
SerpApi's core position is straightforward: no company owns publicly accessible web content simply because it indexes or aggregates it. That argument carried last week. If it holds, platforms seeking to fence off their data from AI scrapers will need a stronger legal theory than the one Google is currently testing.
On balance, one loss does not decide this. Google says it will continue, the DMCA framing remains novel at the appellate level, and the specific question of whether licensed content inside knowledge panels creates a defensible copyright position has not been resolved.