Google has won dismissal of antitrust lawsuits brought by Chegg and Penske Media over AI-generated search summaries. The immediate GEO implication is uncomfortable but clear: a publisher’s loss of referral traffic, revenue, or control over how its indexed content appears in AI answers does not, by itself, currently create much near-term leverage through this particular U.S. antitrust theory. arstechnica.com
That does not mean the underlying commercial problem has disappeared. Judge Amit Mehta explicitly recognized the alleged harms to publishers and other creators whose content Google takes and repurposes without compensation. But he held that courts must apply antitrust law as written rather than use it as a substitute for legislative action over economic harm caused by innovation. theglobeandmail.com
The claim that failed was not merely “AI harms publishers”
Chegg and Penske alleged that Google forced publishers into a choice: permit AI Overviews of their content or lose inclusion in Google’s search index. They said the arrangement diverted traffic from their sites and reduced revenue. enterpriseai.economictimes.indiatimes.com
The central problem, in Mehta’s view, was the supposed exchange. The plaintiffs argued that making content available to Google came with an expectation of search traffic. Mehta rejected that as an enforceable agreement: “But an expectation is not an agreement. It is simply how a general search engine works.” theglobeandmail.com
That distinction materially changes how GEO practitioners should interpret publisher complaints about AI search. A search presence can still be commercially valuable. It can still be associated with traffic, discovery, brand recognition, and conversion opportunities. But these dismissals indicate that the historical practical relationship between crawling, indexing, ranking, and referrals is not necessarily a legal commitment Google must preserve when it changes the search-results experience.
In other words, “we made content accessible and Google used to send visits” is not, on the reported reasoning, enough to establish an antitrust violation when Google surfaces AI-generated answers that reduce the need for a click. The lawsuits failed at the threshold: Mehta said the claims “fail to get out of the starting gate.” enterpriseai.economictimes.indiatimes.com
Why this narrows publishers’ near-term bargaining position
The ruling does not say that Google’s AI search practices are good for publishers. It says the claims presented did not establish an antitrust violation. That is a narrower holding than a declaration that publishers have suffered no harm, but it is still consequential for leverage.
A traffic-loss lawsuit can serve two purposes: potentially obtain a legal remedy, and increase negotiating pressure even before a final judgment. The dismissals weaken both functions for claims built chiefly around the proposition that Google unfairly repurposes indexed content while withholding the referrals publishers expected.
Google’s position was that it has no obligation to index publishers’ content on terms the publishers prefer. Chegg and Penske argued that a competitive market would require Google to pay for republishing their work or using their content to train AI systems. The court did not accept their antitrust framing. enterpriseai.economictimes.indiatimes.com
For a GEO lead inside a publisher, that means litigation should not be treated as a near-term operational answer to declining click-through from AI-mediated search. The supplied reporting does not establish what remedy, if any, could succeed under a different legal theory, nor does it establish whether these particular cases will be appealed. It does establish that this antitrust route has now failed for both Chegg and Penske.
The ruling also follows a similar rejection in March of claims by another publisher suing Google. theglobeandmail.com That reported sequence makes it harder to regard the Chegg and Penske decisions as an isolated procedural setback.
GEO cannot rely on an implied referral bargain
There is a strategic mistake lurking in much AI-search planning: treating citation, inclusion, or crawlability as if they necessarily produce the old exchange of content for visits. Mehta’s reasoning cuts directly against that assumption in legal terms, but it should also change operational language.
For GEO work, distinguish three things that are often collapsed into one:
- Content availability: whether a platform can index or otherwise use material that a publisher makes available.
- Answer visibility: whether the publisher’s material is represented in an AI-generated result.
- Referral value: whether the result sends a user to the publisher’s site in a way that supports the business.
The cases were driven by the alleged gap between the first and third items: publishers said that content made available for search was harvested for AI answers, while traffic fell. arstechnica.com The dismissal means that gap should be treated as a business exposure to measure, not as a legally protected exchange that can be assumed in planning.
This is not an argument that GEO is pointless, or that publishers should disregard how their work appears in AI answers. It is the opposite. When legal leverage based on traffic loss is weak, the discipline of understanding what answer visibility actually delivers becomes more important. Teams need to be exact about whether an AI-search appearance produces a visit, whether it substitutes for a visit, and what commercial outcome remains when a user’s question is resolved before reaching the page.
The court’s sympathy is not a remedy
The strongest passage for publishers may also be the clearest warning against overreading the decision. Mehta wrote that the court did not treat the alleged harms lightly and was not unsympathetic to publishers, journalists, educators, and other online creators facing the consequences of uncompensated repurposing. arstechnica.com
But sympathy did not convert the alleged conduct into an antitrust violation. The judge’s position, as reported, is that economic harm caused by innovation is not necessarily the product of illegal conduct and that lawmakers, rather than antitrust law, would need to address the gap if a new rule is warranted. arstechnica.com
That is the crucial limit for publishers deciding what to expect next. The ruling does not validate the economics of AI Overviews for content producers. It validates neither a payment model nor a workable publisher opt-out within organic search. It simply removes, for now, one proposed route to force those outcomes in U.S. court.
What remains unsettled
The sources leave major questions unanswered. They do not tell us whether a copyright-based challenge would fare differently, whether Chegg or Penske will pursue further action, or what contractual or product changes Google might make. They also do not establish how much traffic any individual publisher has lost, what kinds of pages are most affected, or whether inclusion in an AI answer creates measurable downstream value.
Ars Technica reports that Google is likely to face more AI-search challenges, while characterizing further U.S. court challenges as uphill without new legislation. It also reports that the European Commission is considering related questions, that the UK ordered Google to provide an AI opt-out for sites remaining in organic search, and that Google is exploring direct payment approaches through a pilot reportedly unpopular with publishers. arstechnica.com Those developments may matter, but they do not alter the present U.S. dismissals.
The near-term conclusion is therefore narrower than “publishers have no options.” Publishers may have legal, legislative, commercial, and technical options that these sources do not resolve. The defensible conclusion is that traffic-loss antitrust litigation has become a weaker source of immediate leverage after the Chegg and Penske dismissals. For GEO teams, that raises the value of rigorous measurement and lowers the value of assuming search visibility comes with an enforceable promise of referral traffic.