Google once positioned its AI Overviews as a helpful shortcut. Summaries drawn from web results would sit atop search pages and answer queries directly. Users clicked less. Publishers watched traffic vanish. Now the feature stands at the center of a growing legal storm.
A Munich court ruled this week that Google bears direct responsibility for false statements generated in those overviews. The decision strips away the traditional safe harbor that protects search engines from liability for third-party content. When the AI synthesizes its own answer, the court said, those words belong to Google.
The case involved publishers wrongly linked to fraud by an AI Overview. German judges treated the summary as Google’s own speech. No longer could the company hide behind its role as mere intermediary. Direct infringer. That label changes everything.
But one ruling hardly stands alone. Publishers have filed antitrust suits on both sides of the Atlantic. They argue Google forces them into an impossible choice. Allow your content to train and fuel AI Overviews or disappear from search results entirely. Either path hurts revenue.
Penske Media, owner of Rolling Stone, Billboard and Variety, sued Google in federal court in Washington. The company claims the AI summaries repackage its journalism without permission. Traffic to its sites dropped. Advertising, subscriptions and affiliate income suffered. The Wall Street Journal reported on the suit when it was filed last September. Google moved to dismiss the case in January, calling the claims legally defective in every way, according to a Reuters account.
Chegg, the online education platform, brought its own antitrust action. The company said AI Overviews eroded its visitor numbers and subscriber base. It accused Google of misusing its monopoly in search to dominate adjacent markets. Similar complaints landed in Europe.
The Independent Publishers Alliance filed an antitrust complaint with the European Commission in July 2025. It sought interim measures to halt what it called irreparable harm. Publishers cannot meaningfully opt out of AI use without sacrificing visibility in ordinary search. Reuters covered the filing exclusively.
UK regulators moved faster. The Competition and Markets Authority designated Google with strategic market status. In June it imposed conduct requirements that give publishers more control. They can now prevent their content from powering AI features such as Overviews or from training separate models. The rules do not affect traditional search rankings. Reuters detailed the enforcement.
Yet opt-out remedies carry their own problems. Publishers who step away may see their ideas appear anyway through syndicated sources that stayed in. Summaries could grow less accurate without diverse voices. European Commission investigators opened a formal probe last December. They are examining whether Google provided appropriate compensation and real opportunities to refuse use of content. The BBC reported on the investigation’s focus on both web publishers and YouTube creators.
Copyright questions hover in the background. Ongoing litigation in the United States, including consolidated cases over training data for Gemini, tests fair use boundaries. Authors and illustrators sought class certification earlier this year. They want transparency and consent before their works feed AI systems. But courts have yet to deliver clear verdicts on whether generating summaries from indexed web pages constitutes infringement.
Defamation and misinformation fears feel more immediate. Early AI Overviews famously suggested eating rocks or adding glue to pizza. Health-related summaries later raised alarms. Some provided incorrect and potentially dangerous medical advice. Google pulled certain answers after investigations found factual errors.
The German decision amplifies those worries. At scale, even a 9 percent error rate produces millions of faulty responses daily. Each one now carries potential legal exposure. Analysts on X noted the precedent could apply to any AI answer engine that synthesizes original content, from Perplexity to Claude-powered search tools. One post called it a wake-up call. Another predicted labs would rush to improve source traceability in retrieval-augmented generation systems.
Google maintains the feature improves search. It cites high accuracy rates and continuous refinements. The company points to user satisfaction data and argues Overviews drive discovery that benefits the web overall. In regulatory filings it stresses that AI summaries complement rather than replace links to original sources.
But data from publishers tells a different story. Traffic declines reached 30 to 60 percent for some content categories after AI Overviews rolled out more widely. Independent news sites suffered most. Their advertising models rely on clicks that no longer come.
Competition authorities see broader market distortion. The UK CMA and European Commission worry that Google uses its search dominance to control the next layer of information delivery. By answering questions directly, the company keeps users inside its product. Rivals in search, publishing and even e-commerce lose out.
So far Google has offered concessions. It tested new controls for UK publishers. It promised greater transparency. Whether these steps satisfy regulators remains uncertain. The EU investigation continues. Antitrust complaints multiply. And now a court has opened the door to liability for every hallucination that harms reputation.
The original concerns raised in technology coverage have only grown sharper. TechRepublic examined the legal risks surrounding copyright, defamation and competition issues as the feature expanded. Those risks no longer sit in the realm of theory. They have names, court dockets and regulatory deadlines.
Publishers aren’t waiting for perfect clarity. Some have shifted resources toward direct subscription models and email lists. Others experiment with blocking bots or adding stronger terms of service. A few have joined lobbying efforts aimed at the U.S. Department of Justice. The fight stretches into 2026 and beyond.
For Google the stakes extend past any single lawsuit. Its search business still generates the majority of revenue. AI Overviews were meant to defend that franchise against chat-based competitors. Instead they invited new attacks on multiple legal fronts. Antitrust. Copyright. Defamation. Each carries financial and reputational cost.
Industry watchers expect more rulings like the one from Munich. European courts have shown willingness to treat generative outputs as the responsibility of the deploying company. If higher courts uphold the logic, AI search products worldwide may need to add disclaimers, human oversight or stricter guardrails. The era of frictionless, authoritative-sounding summaries could narrow.
Users notice the change too. Many appreciate instant answers. Others have learned to scroll past the overview and click through for context. Trust erodes when summaries get basic facts wrong or attribute ideas incorrectly. One mistaken label can damage careers or businesses.
The path forward likely involves compromise. Regulators may require genuine opt-out mechanisms that don’t penalize visibility. Google might share more revenue with content creators whose material appears in training or summaries. Courts could draw distinctions between verbatim copying and transformative use.
Yet momentum builds against the status quo. A German court just declared that Google’s words are Google’s responsibility. That simple statement carries weight. It signals to every technology company building similar products that the old rules no longer apply unchanged.
Publishers, regulators and now judges have drawn a line. The question is whether Google, and the industry it leads, will adapt before more precedents harden against them. The legal pressure on AI Overviews shows no sign of easing. And the consequences could reshape how information flows across the web for years to come.
WebProNews is an iEntry Publication