A fresh complaint against SerpApi reached the court on August 10, arriving three weeks on from the dismissal of both Digital Millennium Copyright Act claims Google had brought over SearchGuard, its access control against Search-result scraping. Chief U.S. District Judge Yvonne Gonzalez Rogers issued that July ruling. The new filing landed on the deadline she set: Google had 21 days to fix the pleading gap identified in the ruling. Search Engine Journal’s Matt G. Southern reported the amendment.
The new complaint quotes obligations from Google’s own licensing partners. One partner’s agreements, in versions dating back to 2017, require Google to take commercially reasonable steps to keep licensed content away from outside access. A separate partner’s contract bars its material from being downloaded by any third party. Google’s Reddit agreement, the filing says, similarly bars outside parties from pulling out and reselling licensed material. Reddit pushed for technical safeguards after suspecting scrapers were flipping that content for profit. Google also cites its privacy policy, which promises to protect the images and reviews people submit to Search.
Google frames the filing as satisfying what the court demanded, though it adds a qualifier. If authorization from copyright owners turns out to be required for SearchGuard, the complaint states, Google already had it and still does. That hedge preserves Google’s underlying position that deploying an access control system does not require owner permission at all, an argument Gonzalez Rogers rejected in July.
The July 20 ruling had split the dismissed claims in two. Where SearchGuard protected results with no copyrighted material, Gonzalez Rogers dismissed those claims outright, with no chance to refile, because the statute only covers copyrighted works. Claims tied to results that did contain copyrighted material got leave to amend instead. The judge cited Section 1201(a)(3)(B) of the DMCA, which requires an access control to operate under authorization from the copyright holder. In its pleading Google said it holds licences covering the imagery used in Knowledge Panels, a Search feature that surfaces summary facts alongside results, but supplied no license terms, leaving the judge unable to find Google authorized to run SearchGuard on that material.
The same order rejected several of SerpApi’s other defenses. Gonzalez Rogers found the DMCA lets any injured party sue, not only copyright holders, so Google had standing to bring the case. She also found SearchGuard did not need to target copyrighted material exclusively, and that Google’s circumvention allegations gave no independent ground for dismissal.
The amended complaint narrows SearchGuard’s description to copyrighted works licensed for Search, but not the damages claim. Google still alleges billions of circumvention events without limiting that count to result pages carrying licensed content, even though the order forecloses claims resting on pages with no copyrighted material.
Meredith Rose, a DMCA specialist at the advocacy group Public Knowledge, told Ars Technica after the July ruling that the approach is risky for Google. Making the authorization argument forces the company to disclose how much copyrighted material sits inside its Knowledge Panels. Rose said Google had backed itself into “a corner.”
Whether the surviving claims hold now turns on a second question layered on top of circumvention: do Google’s licensing contracts actually show copyright owners authorized SearchGuard to defend their material? That question pulls private contract terms into a case built on scraping, and it reaches beyond Google and SerpApi.
Most rank trackers, SERP monitoring tools and AI visibility platforms operate by pulling data from Google Search results, the exact activity SearchGuard is built to block. If a licensing clause between Google and a content partner can anchor an anti-circumvention claim, the legal question stops being only about Google’s own terms of service. It becomes a question of third-party rights riding inside the results page, an exposure that touches any monitoring tool a search team already pays for.
SerpApi’s counsel said Tuesday that the company remains confident and intends to respond, the next filing due in the case. The July order bars SerpApi from raising, in a second motion to dismiss, any argument it could have raised the first time. Discovery stays frozen until the court rules on that motion. A parallel case is moving too. Reddit brought its own DMCA claims against SerpApi in the Southern District of New York, and Judge Paul Engelmayer allowed those to go forward on July 31, per MediaPost.
For any team paying for a scraping-based rank tracker or AI-visibility monitor, the licensing question is one to track alongside the docket. If the amended complaint survives a second motion to dismiss, the exposure shifts from Google’s terms of service to the specific content deals embedded inside search results. That shift is worth flagging to legal or procurement now, not after a ruling forces the issue.
Search Engine Journal, in an August 12, 2026 report by Matt G. Southern, first detailed the amended complaint.
