Google's DMCA Scraping Lawsuit Against SerpApi Just Died
A federal judge just told Google that not every anti-bot wall is a copyright fence. On July 20, 2026, the DMCA scraping lawsuit Google filed against SerpApi collapsed — most of it permanently, on the simple ground that ordinary search results aren't a copyrighted work in the first place.
The short version
- Chief Judge Yvonne Gonzalez Rogers of the Northern District of California dismissed Google's DMCA claims against SerpApi on July 20, 2026, in Case No. 25-cv-10826-YGR. (PPC Land)
- Claims covering ordinary search results — links, snippets, index data — were dismissed with no chance to refile, because that data isn't copyrighted and Section 1201 only protects access controls guarding actual copyrighted works.
- Claims tied to Knowledge Panels, where a licensed image sometimes appears, got a narrower lifeline: Google has 21 days to file a new complaint alleging it actually authorized SearchGuard to protect that specific content.
- Google's own December 19, 2025 complaint had sought statutory damages of $200 to $2,500 per act of circumvention against a company it said grew query volume 25,000% in two years — a number that could have dwarfed SerpApi's reported annual revenue of a few million dollars.
- This is the first reasoned court ruling on whether an anti-bot layer over public search results clears the DMCA's copyright bar. The answer, for the general case, is no. (SerpApi)
What the Court Actually Decided
Google sued SerpApi, an Austin-based scraping API with about 42 employees, under two DMCA Section 1201 theories. One claimed SerpApi circumvented SearchGuard, Google's anti-bot system launched in January 2025. The other claimed SerpApi trafficked in a service built to do that circumventing.
Judge Gonzalez Rogers split her order in two. Where SearchGuard blocked access to plain search results with no copyrighted content, she dismissed the claims with prejudice — Google can't try again on that theory. Where a result page included a Knowledge Panel with a licensed image, she dismissed but left the door open, giving Google 21 days to plead facts it hadn't pled the first time.
Does This Ruling Make Web Scraping Legal?
No — it makes one narrow point: an anti-bot system doesn't automatically count as DMCA copyright protection just because it blocks bots. The court still found Google adequately alleged that SerpApi circumvented SearchGuard; it just ruled that circumvention alone isn't a DMCA violation unless the thing being guarded is a copyrighted work.
That distinction matters more than it sounds. Plenty of anti-scraping arguments assume that beating a technical barrier is itself illegal. This ruling says the barrier has to be protecting something copyrightable before DMCA liability even enters the conversation.
Why SearchGuard Failed the Copyright Test
Google's own pleading did the damage. It described search results as compilations of public web data, arranged by relevance, "often" accompanied by a Knowledge Panel that "may" contain copyrighted content. Judge Gonzalez Rogers read that as an admission: some results carry copyrighted material, most don't.
Section 1201 only reaches measures that "effectively control access to a work protected under" the Copyright Act. If SearchGuard sits in front of a page with no copyrighted content, it isn't protecting a copyrighted work — so DMCA's anti-circumvention rule can't apply to that page, full stop.
A second problem sank even the Knowledge Panel claims for now. Section 1201(a)(3)(B) requires that the access control operate "with the authority of the copyright owner." Google alleged it licensed the images but never pled the terms of those licenses, so the court couldn't confirm the copyright owners actually authorized SearchGuard to guard their content.
| Claim track | Ruling | Can Google refile? |
|---|---|---|
| Ordinary search results (no copyrighted content) | Dismissed as a matter of law | No — with prejudice |
| Knowledge Panel results (licensed images present) | Dismissed for missing facts on authorization | Yes — within 21 days |
What This Means If You Build on Public Data
The ruling doesn't touch every scraping dispute — Reddit's parallel DMCA suit against SerpApi, Oxylabs, AWMProxy, and Perplexity AI, filed October 22, 2025 in the Southern District of New York, also leans on SearchGuard as an access control, and that case is still live. SerpApi has moved to dismiss it on overlapping grounds. Nobody should read one district court order as settled law across every jurisdiction and every anti-bot system.
What it does establish is a template other courts will now measure new DMCA-over-scraping claims against: describe your access control, then show it protects something actually copyrighted, with your authority to protect it. A platform operator who wants Section 1201 power over scrapers now has to plead specifics, not just point at a bot wall and call it copyright infringement.
For anyone whose product depends on reaching public pages reliably — which is the exact problem ScrapeOps sits in front of — the practical takeaway isn't "scraping is now safe." It's that the legal exposure for reaching a page and the technical difficulty of reaching it are two separate problems, and this ruling only moved the first one. The second is still a daily operational fight against CAPTCHAs, IP blocks, and JavaScript challenges like SearchGuard itself, regardless of what a court says about copyright. We track that operational side of the fight in our dispatches, including Cloudflare's new AI crawler rules that took effect the same month this ruling landed.
The Bigger Pattern
This case sits inside a year of Google tightening access to its own search data. It removed the num=100 parameter in September 2025, forcing SEO tools into ten requests where one used to work, then blocked a workaround within five days of it shipping. It also throttled SerpApi's own fallback API to three results per query around the same time.
None of that required a courtroom. A DMCA win would have let Google reach past its own product surface and into contract-adjacent territory — treating unauthorized access to public pages the way copyright law treats unauthorized copying of a movie. That's the theory the court rejected, at least for the ordinary case, and it's why a ruling about search-results scraping has weight for anyone building on any public dataset, not just SEO tool vendors. For more on how access rules and data rights are shifting across the industry, see our research page.
Frequently Asked Questions
What is Google's SearchGuard system? SearchGuard is an anti-bot system Google launched in January 2025 that sends a JavaScript challenge to search queries from unrecognized sources. Human browsers solve it without friction; automated systems querying at scale typically can't, and get denied access.
Why did Google's DMCA lawsuit against SerpApi get dismissed? Judge Yvonne Gonzalez Rogers ruled that Section 1201 of the DMCA only protects access controls guarding copyrighted works. Because Google's own complaint described search results as largely uncopyrighted public data, SearchGuard couldn't qualify as a valid DMCA access control for most of the case.
Can Google still win part of its case against SerpApi? Yes, narrowly. Google has 21 days from the July 20, 2026 order to file an amended complaint limited to Knowledge Panel results that contain licensed, copyrighted images — but only if it can plead facts showing it authorized SearchGuard to protect that specific content.
Does this ruling make all web scraping legal under the DMCA? No. It establishes that an anti-bot system isn't automatically a copyright access control — the operator must show it protects a copyrighted work with the owner's authority. Other scraping disputes, including Reddit's separate suit against SerpApi and others, remain unresolved.
Abhishek Gupta is Co-Founder at Dekrypt Labs, building ScrapeOps — the data acquisition engine that turns any question into clean, deduplicated, comprehension-ready sources. dekryptlabs.com