A federal judge threw out Google’s Digital Millennium Copyright Act case against SerpApi, ruling that scraping public search results is not copyright circumvention when those results carry no copyrighted material. SerpApi’s application programming interface feeds the rank trackers, keyword platforms, and competitive-intelligence dashboards that SEO practitioners check daily. The ruling is the first reasoned decision in the US to hold that scraping is not DMCA circumvention when the results carry no copyrighted content. Chief Judge Yvonne Gonzalez Rogers of the US District Court for the Northern District of California signed the order July 20, 2026. SerpApi disclosed the ruling the next day, and the win is narrower than the company’s own framing suggests.
Google filed the case, Google LLC v. SerpApi, on December 19, 2025, and Gonzalez Rogers split her order into two tracks. Claims covering ordinary search results, the ones with no copyrighted content, were dismissed without leave to amend. The judge found those claims defective as a matter of law, leaving Google no path to refile. Claims tied to a copyrighted image inside a Knowledge Panel were dismissed with leave to amend, and Google now has 21 days to file again. The case is not closed.
The order’s central line explains why: “To the extent that Google Search results do not contain any copyrighted content, SearchGuard cannot be said to effectively control access to a work protected under the Copyright Act.” Google’s own complaint supplied the evidence against itself. It described its results as a compilation of publicly available material sorted by relevance. It also said a Knowledge Panel is “often” present and “may” contain copyrighted content. Google never alleged the panel appears on every page, or that it always holds protected material.
A second, independent flaw sank the Knowledge Panel claims too. Section 1201(a)(3)(B) requires that an access-control measure operate “with the authority of the copyright owner.” Google never pleaded facts showing that the owners of images displayed in Knowledge Panels authorized SearchGuard to guard them. Citing the Ninth Circuit’s Blizzard precedent, the court held that Google’s licensing arrangements alone did not satisfy that requirement without pleaded terms.
Google did not lose on every front. The court rejected SerpApi’s argument that the DMCA protects only copyright owners, holding that its zone of interests covers anyone injured by a circumvention violation. It also found Google adequately alleged that SerpApi circumvented SearchGuard, describing automated queries disguised to look human and a solved challenge passed along to browsers that had not solved it. The court further rejected SerpApi’s claim that a 1201(a)(2) violation requires selling circumvention technology outright. Google lost on what SearchGuard actually protects, not on whether scraping happened or who may sue over it.
SearchGuard, launched in January 2025, challenges unfamiliar traffic with a JavaScript test that ordinary browsers solve automatically and that automated scraping systems typically cannot pass. Google sought statutory damages between $200 and $2,500 for each act of circumvention. PPC Land reported that figure, applied to SerpApi’s query volume, threatened to dwarf the few million dollars in annual revenue the company reports.
The suit followed a year in which Google tightened the SERP-data pipeline through other means. It removed the num=100 parameter in September 2025, a change Semrush said raised the cost of full-SERP analysis roughly tenfold. Google also blocked a DataForSEO workaround within five days and cut SerpApi’s own Light Fast API to three organic results per query.
SerpApi chief executive Julien Khaleghy called the outcome “a win not just for SerpApi, but for all who depend on an open internet.” He added, “We’re pleased that the court rejected Google’s attempts to expand the DMCA to assert control over access to public pages.” Google has not commented on the ruling.
For SEO teams, this ruling does not make SERP scraping free of legal risk, and it does not resolve the Knowledge Panel claim, which Google can revive within three weeks. It does draw a boundary no US court had drawn before: a results page with no copyrighted content cannot be locked behind the DMCA’s anti-circumvention provisions. Paired with the post-num=100 cost increase, that boundary sets the terms for the next round between Google and every rank-tracking vendor whose data SEO teams buy.
Per PPC Land, which reported the ruling on July 21, 2026, citing a court order Chief Judge Yvonne Gonzalez Rogers signed July 20, 2026.
