# Google’s DMCA Case Against SerpApi Narrows After Court Rejects Claims Over Uncopyrighted Search Results > Source: > Published: 2026-07-28 09:22:41.810513+00:00 # Google’s DMCA Case Against SerpApi Narrows After Court Rejects Claims Over Uncopyrighted Search Results - A federal judge permanently dismissed Google’s DMCA claims where its SearchGuard system restricted access to search results containing no copyrighted material. [[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) - Google has until August 10, 2026, to amend narrower claims involving copyrighted components, including licensed images displayed in Knowledge Panels. [[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[2]](https://arstechnica.com/tech-policy/2026/07/google-wont-give-up-odd-war-against-ai-web-scraping-despite-court-loss/) - The ruling does not declare SerpApi’s practices lawful generally or decide whether scraped material may be used for AI training. [[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) - Reddit’s related case against SerpApi, Perplexity and other scraping providers remains pending after nearly three hours of oral argument on motions to dismiss. [[5]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf)[[7]](https://news.bloomberglaw.com/business-and-practice/reddit-defends-authority-to-sue-perplexity-over-data-scraping) A federal judge has blocked Google from using the Digital Millennium Copyright Act to pursue SerpApi for bypassing anti-bot controls protecting ordinary search results, narrowing one emerging legal strategy for restricting automated access to data used by AI services, search tools and other businesses.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) U.S. District Judge Yvonne Gonzalez Rogers dismissed Google’s two DMCA claims on July 20 in Google LLC v. SerpApi LLC, No. 4:25-cv-10826. She permanently rejected the claims to the extent they concerned search results containing no copyrighted material, finding that the DMCA provisions Google invoked apply only when a technological measure controls access to a work protected by copyright.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) The ruling leaves Google a smaller path forward. Rogers allowed the company 21 days to amend claims involving copyrighted components that can appear in search results, including images licensed for Knowledge Panels. Google spokesperson José Castañeda said the company plans to amend and welcomed the court’s rejection of most of SerpApi’s other legal arguments.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[2]](https://arstechnica.com/tech-policy/2026/07/google-wont-give-up-odd-war-against-ai-web-scraping-despite-court-loss/) ## SearchGuard and the limits of the DMCA Google sued SerpApi in December 2025 under Sections 1201(a)(1)(A) and 1201(a)(2) of the DMCA. Those provisions prohibit circumventing access controls protecting copyrighted works and trafficking in products or services designed for that purpose. Google did not bring a conventional copyright-infringement claim.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[4]](https://blog.google/innovation-and-ai/technology/safety-security/serpapi-lawsuit/) According to Google’s complaint, SerpApi sends automated requests to Google Search and sells the resulting information through an API. Google alleged that SerpApi masked automated queries as human traffic and reused successful responses to SearchGuard, a JavaScript challenge system introduced in January 2025. Google said the activity generated billions of automated queries and imposed infrastructure and security costs. Those allegations have not been tested at trial.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[4]](https://blog.google/innovation-and-ai/technology/safety-security/serpapi-lawsuit/) Rogers divided the claims according to the content behind SearchGuard. Google could not use Section 1201 to protect search output consisting of uncopyrightable facts or otherwise unprotected information, she found. That portion was dismissed without leave to amend because the defect was legal rather than merely a lack of detail.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) The judge reached a different conclusion for results containing copyrighted components. She rejected SerpApi’s argument that only a copyright owner or exclusive licensee can sue under the DMCA, finding that the statute permits a person injured by a violation to bring a claim. Google’s pleading still failed because it did not allege facts showing that the owners of licensed content had authorized SearchGuard to control access to their works.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) The order did not accept SerpApi’s argument that the alleged methods could not qualify as circumvention. At the pleading stage, Google’s account of masking automated traffic and sharing successful challenge responses was sufficient. Google must file any amended complaint by August 10. Discovery is stayed until an amended complaint, if any, and a resulting motion to dismiss are resolved. No injunction is in place.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[8]](https://searchengineland.com/google-loses-key-dmca-claims-against-serpapi-in-scraping-lawsuit-483185) ## SerpApi claims a win for public-web access SerpApi called the dismissal a victory for companies and researchers that depend on public information. The company said the court rejected an attempt to expand the DMCA into a mechanism for controlling access to public pages and said it would continue serving developers, researchers, businesses and AI companies.[[3]](https://serpapi.com/blog/google-v-serpapi-the-court-granted-our-motion-to-dismiss/) Google argues that SerpApi’s practices differ from ordinary web crawling. The company says SerpApi evades technical controls, disguises traffic and resells licensed images and real-time information without respecting instructions from websites and rights holders.[[4]](https://blog.google/innovation-and-ai/technology/safety-security/serpapi-lawsuit/) The July 20 order does not establish that scraping publicly viewable information is always lawful. It resolves two DMCA theories at the pleading stage and preserves a possible claim when an access control protects identifiable copyrighted material with the copyright owner’s authority. It does not decide whether SerpApi infringed copyright, violated a contract or acted unlawfully under another statute.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) ## Reddit case tests a broader theory The decision arrived as SerpApi defends a separate case brought by Reddit in the Southern District of New York. Reddit’s amended complaint names SerpApi, Perplexity, Oxylabs and AWMProxy and asserts three DMCA claims, along with unfair-competition, unjust-enrichment and civil-conspiracy claims. Reddit seeks damages and an injunction restricting access to Reddit data through both Reddit and Google.[[5]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf) Reddit alleges the three scraping providers bypassed Google’s controls to retrieve more than 3.1 billion search-results pages containing Reddit material during a two-week period in July 2025. Its complaint describes proxy rotation, forged or manipulated credentials and high-volume requests designed to appear to come from individual users. The figures and descriptions are Reddit’s allegations, based partly on information it says it obtained through a subpoena to Google; they are not judicial findings.[[5]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf) Reddit also says it created a test post containing an unusual hexadecimal string that was served only to Google’s crawler. The string later appeared in a Perplexity response, which Reddit argues shows that its material traveled through Google search results into Perplexity’s system. SerpApi counters that it accessed Google’s publicly available results rather than Reddit’s servers, that Reddit users retain ownership of their posts and that the snippets at issue generally are not copyrightable.[[5]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf)[[6]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/SerpApi-s-Motion-to-Dismiss-First-Amended-Complaint---Reddit-v.-SerpApi.pdf) Judge Paul Engelmayer heard nearly three hours of arguments on July 23 over motions to dismiss the Reddit case. Bloomberg Law reported that the judge questioned the extent of Reddit’s authority to sue over copyrights held by its users. The motions remained pending after the hearing, and no injunction had been entered.[[7]](https://news.bloomberglaw.com/business-and-practice/reddit-defends-authority-to-sue-perplexity-over-data-scraping) The California ruling does not bind the New York court, and Reddit’s complaint is broader than Google’s two-count case. But the same threshold problem is now central to both disputes: an anti-bot system is not automatically a copyright access control merely because it restricts automated traffic.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf)[[5]](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf) For AI-data disputes, Rogers’ order draws a limited boundary. Platforms invoking the DMCA must identify protected material and plausibly allege that the relevant control operates with the copyright owner’s authority. The order does not decide whether publicly available material can be copied for model training, and the Google case itself concerns access to search results rather than proof that SerpApi trained an AI model.[[1]](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) ## Companies mentioned ## Further sources [[1] U.S. District Court for the Northern District of California, Google LLC v. Serp… ↗](https://chatgptiseatingtheworld.com/wp-content/uploads/2026/07/Google-v-SerpApi-order-granting-motion-to-dismiss-July-20-2026.pdf) [[2] Ars Technica, “Google and Reddit do not own the Internet,” web scraper says aft… ↗](https://arstechnica.com/tech-policy/2026/07/google-wont-give-up-odd-war-against-ai-web-scraping-despite-court-loss/) [[3] SerpApi CEO Julien Khaleghy, “Google v. SerpApi: The Court Granted Our Motion t… ↗](https://serpapi.com/blog/google-v-serpapi-the-court-granted-our-motion-to-dismiss/) [[4] Google General Counsel Halimah DeLaine Prado, “Why we’re taking legal action ag… ↗](https://blog.google/innovation-and-ai/technology/safety-security/serpapi-lawsuit/) [[5] U.S. District Court for the Southern District of New York, Reddit Inc. v. SerpA… ↗](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/Reddit-s-First-Amended-Complaint---Reddit-v.-SerpApi.pdf) [[6] SerpApi’s motion to dismiss Reddit’s first amended complaint, filed March 13, 2… ↗](https://storage.ghost.io/c/a5/00/a5004977-0dd2-4bcd-9292-dd0e05d4c59e/content/files/2026/03/SerpApi-s-Motion-to-Dismiss-First-Amended-Complaint---Reddit-v.-SerpApi.pdf)+2 more The stories that matter, in one email. Free — unsubscribe anytime.