A federal court last week dismissed Google's Digital Millennium Copyright Act lawsuit against web scraper SerpApi, finding Google lacks standing to sue over content it does not own. The judge gave Google 21 days to amend its complaint, and Google confirmed it will refile, narrowing the case to material inside its search knowledge panels that it directly licenses from rights holders. The ruling lands as Reddit pursues a nearly identical DMCA claim against SerpApi and Perplexity filed in October 2025.
Google sued SerpApi in December 2025, accusing the company of circumventing anti-scraping technology and reselling scraped Google search data through an unauthorized API service. The theory was unusual because search results themselves cannot be copyrighted. Google argued the anti-scraping tech protected copyrighted material licensed from third-party rights holders that appears inside knowledge panels, and that SerpApi's activity threatened its ability to profit from billions of bot searches.
Reddit's earlier case leaned on a similar structure, alleging SerpApi evaded two levels of security: Reddit's own scraping controls plus Google's controls protecting Reddit content in search results. Google cited Reddit's suit when announcing its own, calling the action a last resort against malicious scraping. SerpApi has framed both suits as attempts to retroactively assert control over content the platforms neither authored nor own.
Key facts
- 01A federal court dismissed Google's DMCA lawsuit against SerpApi last week, ruling Google lacks standing because it doesn't own the copyrighted content in search results.
- 02Google was given 21 days to amend its complaint and has confirmed it will refile, narrowing the fight to licensed content in knowledge panels.
- 03Reddit filed a parallel DMCA suit in October 2025 against SerpApi and Perplexity; Google filed in December 2025 citing Reddit's action.
- 04SerpApi customers include Nvidia, Uber, and Adobe, who rely on structured access to search data.
- 05Reddit alleged SerpApi evaded two levels of security, its own scraping controls plus Google's scraping controls on Reddit content.
SerpApi told Ars Technica it is prepared to keep fighting despite the cost. The company said its customers, including Nvidia, Uber, and Adobe, depend on structured search access for legitimate business use and have faced uncertainty as the litigation has dragged on.
“The bottom line is that both Google and Reddit appear to be engaged in attempts to use the DMCA to wall off the open Internet by retroactively claiming control over content that they didn't author and don't own.”— SerpApi, web scraping company
Meredith Rose, senior policy counsel at Public Knowledge, said Google and Reddit are grasping at whatever tool is available in response to three years of accelerating AI scraping. She called the DMCA argument bizarre and outside what the law contemplated, but not surprising given how effectively the statute has historically forced licensing conversations.
The dismissal is unusually early. Rose noted that motions to dismiss are rarely granted at this stage, and the court's reasoning came down to a simple gap: Google did not allege enough about what it was actually protecting that qualified as copyrighted. Under the DMCA, standing requires the plaintiff to be a copyright owner, an exclusive licensee, or the entity deploying the protection measure on behalf of one.
Rose said the same logic looks bad for Reddit, which faced its own hearing on SerpApi's motion to dismiss last Thursday. Reddit is not the copyright owner or exclusive licensee of user-generated content that appears in Google search results, which is the category of material at issue. SerpApi said Reddit did not appear in the courtroom, and that the judge seemed focused on whether Reddit's agreement with Google actually authorized Google to protect Reddit's content.
“They're in the right from a policy perspective here.”— Meredith Rose, Senior Policy Counsel, Public Knowledge
Google's path forward is narrow. If it can show that specific rights holders directly authorized Google to deploy anti-scraping technology to protect their licensed content inside knowledge panels, it may salvage a limited claim. But Rose said Google has talked itself into a corner. Arguing knowledge panels are full of copyrighted material would invite scrutiny of the unlicensed portions, potentially opening Google to fair use fights it does not want.
Google spokesperson José Castañeda said the company was pleased the court rejected nearly all of SerpApi's other legal arguments and disputed the standing challenge.
“We look forward to filing an amended complaint, as the Court invited us to do, and we remain committed to protecting our services and partners from unauthorized access.”— José Castañeda, Google spokesperson
SerpApi says the broader stakes go beyond its own customers. The company argues that platform attempts to weaponize the DMCA threaten all users of the open web, and that a Google win would set a precedent for retroactively enclosing public data. Rose agreed on the policy substance, saying that while SerpApi has not endeared itself to everyone, its position aligns with the open web's interests.
Rose traces the current fight to the summer of 2023, when what she called an API apocalypse began. Publishers have spent the past three years cutting off access, throttling crawlers, and hunting for legal tools to stop high-volume AI scraping. The result, she said, has been a re-enclosure of a large portion of the web that harms not just AI training but also research, archiving, journalism, and public health reporting that depend on automated crawling at scale.
The immediate question is whether Google can craft an amended complaint that survives a second motion to dismiss without exposing its knowledge-panel construction to fair use challenges. The deeper question, for the AI market, is whether the DMCA becomes a workable choke point on scraping at all. If courts continue to reject standing claims from platforms that host but do not own content, AI companies and data intermediaries gain a stronger footing to keep crawling public pages. If Google finds a workable theory, every large platform with a licensing agreement gains new leverage over the scrapers feeding model training and retrieval pipelines.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
