The ruling that just landed
Here's the short version: on July 20, 2026, a federal judge in San Francisco granted SerpApi's motion to dismiss Google's entire lawsuit. Full stop. The court rejected Google's attempt to weaponize the DMCA against a company that simply builds APIs for publicly visible search results.
SerpApi CEO Julien Khaleghy called it "a win not just for SerpApi, but for all who depend on an open internet." Fair enough. But the real story here isn't about one Austin-based startup beating Google at its own game — it's about what this ruling says about the boundaries of copyright law in an era where everyone wants to control how data flows.
Google now has 21 days to amend its complaint. That clock is ticking.
This decision directly impacts AI in mental health care, where researchers and tools rely on public search data to identify trends, assess stigma, and develop early-intervention models. If Google’s theory had prevailed, even scraping public mental health forums or symptom reports could have been legally risky.
How we got here
Google filed suit against SerpApi on December 19, 2025. No warning shot. No cease-and-desist. No phone call. Just a complaint in the Northern District of California, assigned to Judge Yvonne Gonzalez Rogers.
SerpApi's general counsel, Chad Anson, joined the company days before the lawsuit hit. His opening move was to file a motion to dismiss — and he argued something that should have been obvious from the start: Google's "SearchGuard" isn't an access control in any meaningful DMCA sense.
Here's what Anson laid out: SerpApi provides programmatic access to search results that are publicly visible to anyone using a standard web browser. No authentication bypassed. No private accounts accessed. No non-public data retrieved. The API responses mirror how those results appear on search pages, complete with links and citations back to original sources.
Google's theory was that its SearchGuard protocols should get the same legal status as logins, passwords, and encryption. Anson pushed back hard. SearchGuard, he wrote, is "a euphemism for its traffic management tools." It doesn't protect copyrighted works from being copied. It manages how much traffic Google gets.
The court agreed.
Why the DMCA angle was always a stretch
The Digital Millennium Copyright Act was written in 1998. It was designed to protect encrypted DVDs from piracy and software from unauthorized copying. Not traffic management.
Google tried to stretch Section 1201 of the DMCA into something it was never meant to be — a tool for controlling access to publicly visible web content. The problem? Publicly visible information isn't the same as a locked door.
As Anson put it in SerpApi's blog post: "Treating [traffic management] as copyright-enforcement mechanisms would stretch the law far beyond its intended purpose, threaten anyone who wants access to public data, and create a dangerously reckless lane for future suppression of innovation."
That's the core of it. If Google wins this argument, every website could layer on traffic controls and then invoke copyright law to restrict who accesses their public content. The precedent would be enormous — and deeply troubling for developers, researchers, and anyone building on top of publicly available information.
For AI in mental health services, this means continued freedom to analyze public sentiment, track crisis keywords, and support community-based monitoring tools without fear of legal retaliation. The ruling preserves the legal foundation for these vital applications.
The antitrust irony
This is where it gets really interesting.
Google's own antitrust case just established that the company holds and unlawfully maintains monopoly power in general search. In that litigation, Google emphasized how open its search engine is — how users can access results without logging in, how that openness supports competition.
Now Google is suing to restrict access to the exact same publicly displayed information when others rely on it to build competing systems.
Rules for thee, not for me. The hypocrisy is almost too clean to be real.
Anson called it out directly: "Google built its business by systematically accessing, indexing, and organizing publicly available information across the web. By its own admission, this kind of access is central to the operation of search. Yet suddenly, in its allegations against us, Google advances the opposite position."
When you're a monopolist, "open access" is great — until someone else uses it to compete with you.
Who else is watching
The ruling landed on July 20. By July 21, Reddit had already cited it in its own case — a separate lawsuit in the Southern District of New York (No. 1:25-cv-08736). Reddit filed a Notice of Supplemental Authority pointing directly to the Google v. SerpApi dismissal order.
That tells you everything about the broader stakes.
This isn't just about one API provider. It's about AI developers building systems that need real-time information. Security researchers investigating exposed systems. Academic researchers studying misinformation, market dynamics, search behavior. Accessibility tools helping users interact with content in new ways.
If Google's theory had survived, the cost of accessing public data would have gone up. The legal uncertainty would have chilled innovation across the board.
As Khaleghy wrote: "SerpApi will continue supporting developers, AI companies, researchers, and businesses that rely on access to public search information." And now, with a federal court's backing, they've got more leverage than ever.
Learn more about how AI in mental health care—opportunities and risks beyond large language models is shaping public health infrastructure.
What happens next
Google has 21 days to amend its complaint. That window closes around August 10, 2026.
Here's what I'm watching for: Will Google try to reframe its claims in a way that actually fits within DMCA's intended scope? Or will they double down on the traffic-control argument and risk another dismissal?
The court's order is available as a PDF — SerpApi published it on their blog. Judge Gonzalez Rogers was clear: Google's allegations weren't sufficiently supported. That's a pretty harsh assessment for a company that spends hundreds of millions on litigation every year.
If Google amends successfully, this case isn't over. But the bar just got a lot higher.
For now, SerpApi is breathing easier. Their customers — developers, AI companies, researchers — can keep building. The internet stays a little more open than it would have been if Google had won.