Google vs SerpApi: The lawsuit reshaping AI and scraping war

A large-scale cleanup operation

The complaint filed by Google on December 19, 2025, before the U.S. District Court for the Northern District of California (case number 25-10826) pulls no punches in characterizing SerpApi's activities as a parasitic business model. The figures cited by Mountain View in an article published on its official blog are staggering and illustrate the scale of data extraction occurring daily on its servers. According to the legal document, the artificial traffic generated by SerpApi increased by 25,000% within two years.

This exponential growth is no accident. It corresponds to the surging demand for fresh data to feed third-party applications. Google accuses the Texas-based company of sending billions of automated queries, unnecessarily overloading its infrastructure without paying a single cent in compensation. The tech giant highlights a fundamental asymmetry: SerpApi commercializes access to data it does not produce, while leaving Google to foot the bill for hosting and processing requests.

The accusation also concerns the method. To achieve such volumes, SerpApi goes beyond simply reading web pages. The company deploys considerable ingenuity to mimic human behavior, systematically circumventing CAPTCHA and Google's security devices. It is precisely this technological circumvention that allows Google to invoke the Digital Millennium Copyright Act (DMCA), transforming a terms-of-service problem into a federal offense.

Cutting off oxygen to competing language models

It would be naïve to read this case solely through the lens of intellectual property protection or server load management. The timing of this lawsuit reveals an obvious strategic intention to destabilize the supply chain of generative artificial intelligence. Language models like those developed by OpenAI have an Achilles heel: they need real-time web access to remain relevant to current events. Lacking their own mature search index, these actors outsource this connection to reality. SerpApi is known to be a major supplier feeding ChatGPT's web search features. By attacking this supplier, Google strikes its primary competitor indirectly.

The maneuver resembles logistical attrition warfare. Google had already attempted to curb this data hemorrhage through technical means, notably by removing the search parameter num=100, which allowed retrieving one hundred results per page. This modification forced scrapers to multiply requests to obtain the same information volume, mechanically increasing their operational costs. These costs are then passed on to end customers, including AI giants.

The shift from technical countermeasures to legal action indicates that software barriers no longer suffice. Scrapers adapt faster than Google engineers can block them. The legal response aims to create an existential risk for these intermediaries, making Google data supply uncertain and legally toxic for businesses depending on it.

The failure of technical defense against reverse engineering

This lawsuit highlights the extreme sophistication reached by scraping actors. The complaint details a particularly intense cat-and-mouse game. In January 2025, Google deployed a major upgrade to its defensive arsenal, internally dubbed "SearchGuard." This system analyzes behaviors, detects suspicious IP addresses, and imposes humanity verification checks. According to Google's attorneys, SerpApi reacted almost instantaneously, developing circumvention technologies to neutralize SearchGuard. The document cites specific techniques: spoofing location data and software information (User-Agents), but especially "authorization token syndication." Concretely, SerpApi allegedly uses validated sessions on legitimate machines to open access on unauthorized machines.

Google's legal argument rests on Section 1201 of the DMCA, which prohibits circumventing technological measures controlling access to protected works. Relying on this provision, Google claims statutory damages ranging from $200 to $2,500 per circumvention act. Multiplied across billions of requests, the theoretical amount is incomprehensible and would sign the immediate death warrant of any company. Beyond this, the search engine attempts to establish precedent that would criminalize the technical act of scraping whenever it involves bypassing a security barrier, however minimal. If the court follows this reasoning, the entire data and scraping industry faces existential danger.

An alarm signal for the SEO and data ecosystem

Google knows well that SerpApi, with its estimated revenue of just a few million dollars, likely won't have deep enough pockets to pay the colossal damages that could be awarded. The objective lies elsewhere. This is about making an example and sending a clear message to the entire market: the age of impunity is over.

This scorched-earth strategy legitimately concerns the entire SEO and competitive intelligence ecosystem. Many market analysis tools, rank tracking platforms, or price monitoring services rely on technologies similar to SerpApi's. If precedent confirms the illegality of SERP page scraping, entire swaths of the digital economy will need to reinvent themselves or disappear.

Google's public announcement of this complaint on its official blog participates in this psychological warfare. By exposing SerpApi's practices and labeling them as deceptive - particularly the use of the term "Google Search API" implying non-existent official status - the company seeks to discourage potential customers. Using companies might fear having their data access cut off overnight or being accused of complicity.

Data sovereignty as a new strategic asset

This case is part of a broader movement in which raw data becomes a closely guarded and monetizable asset. The era of the open web, where public information was freely accessible and exploitable by all, is giving way to a web of licenses and contracts.

Google applies to others what it practices elsewhere. The company recently signed an expensive agreement with Reddit to access its real-time data. The logic is relentless: if Google must pay to index other platforms' content, why should it let third parties loot its own index for free? Google's added value lies in its capacity to organize global information. Allowing third parties to extract this structured information to resell it amounts to offering them the search engine's added value without them bearing the costs.

The lawsuit against SerpApi prefigures a future where search data access is binary: either through official and costly partnerships, or through increasingly risky technical clandestinity. For AI developers and software publishers, dependence on Google data becomes a major strategic liability that must be quickly secured or circumvented.

Reference sources:

• Why we're taking legal action against SerpApi's unlawful scraping, Google Blog
• Official Complaint