Nearly 300 French newspapers accuse Google of breaking a legally binding pay-for-news pledge, this time over AI Overviews

Nearly 300 French daily newspapers filed a complaint with France's competition authority on Aug. 11, accusing Google of violating a legally binding commitment to negotiate separately with publishers before deploying new AI search features. The complaint, filed by the Alliance de la Presse d'Information Generale (APIG) -- whose roughly 300 members include Le Figaro, Le Monde, Ouest-France, Le Parisien, Les Echos and L'Equipe -- centers on a specific date: Google launched AI Overviews and AI Mode in France on July 22 without first opening a separate negotiation over their use of publisher content, which APIG says breaches commitments Google made legally binding in 2022. Google disputes the characterization. The company told French outlet Franceinfo that it has reassessed its payment system for French publishers "so as properly to take into account the display of protected content within AI Overviews and AI Mode," and that "website publishers can choose whether or not to appear in these AI features."

APIG's central number comes from Arcom, France's own audiovisual and digital regulator: an estimate that AI-generated summaries have already cut traffic to publisher websites by 33% to 38%. That is a regulator's estimate, not a court finding or an admission from Google, and it measures overall AI-summary traffic diversion rather than isolating what Google's July 22 France launch specifically caused -- the two are related but not the same claim, and APIG's complaint leans on the larger, regulator-sourced number to establish that the stakes are real, not to prove Google's July rollout alone caused a 38% drop.

This is not a new fight -- it is the same fight, with AI Overviews as the newest trigger. The underlying legal framework dates to the EU's 2019 Copyright Directive, which created "neighbouring rights": a requirement that large digital platforms negotiate payment terms with publishers before reusing snippets of their content. France's competition authority has enforced that requirement against Google twice already. In July 2021, it fined Google 500 million euros for failing to negotiate "in good faith" with French publishers. In response, Google made a set of five-year commitments in 2022 -- binding promises about good-faith negotiation, transparency, and protection against retaliation through reduced search visibility. In March 2024, the authority fined Google 250 million euros again, this time specifically citing unauthorized use of press articles to train AI systems, among other violations of the same 2022 commitments. APIG's Aug. 11 complaint alleges that Google broke those same 2022 commitments a third time -- by launching AI Overviews and AI Mode without the separate negotiation the commitments require.

Two separate proceedings are now running against Google over essentially the same underlying conduct, and they are easy to conflate. APIG's complaint is a French national case, filed with France's own competition authority, built on France-specific commitments Google made to French regulators. The European Commission's investigation, opened in December 2025 after a separate complaint from the European Publishers Council, is a distinct, EU-wide case under European competition law, covering all 27 member states rather than France alone. A finding in one does not automatically produce a finding in the other, though a Commission ruling against Google's EU-wide practices would likely reshape the French case too.

Google's defense rests on two claims: that it already pays French publishers for AI Overviews use, under a payment system it says it has "reassessed" to account for the feature, and that publishers can opt out entirely. The second claim is not just an assertion -- since June 2026, following pressure from the UK's Competition and Markets Authority, Google has offered publishers a dedicated Search Console toggle that excludes a site from AI Overviews, AI Mode and AI-powered Discover specifically, without removing it from ordinary Search results or the regular Discover feed, and Google says the toggle carries no ranking penalty outside its AI features. That resolves what had been the sharpest version of the opt-out complaint elsewhere -- that refusing AI reuse meant disappearing from Search entirely. It does not resolve APIG's actual claim, which is narrower and predates the toggle question: that no separate negotiation over payment preceded the July 22 France launch at all, which is the specific process the 2022 commitments require regardless of whether an opt-out button exists.

Brussels was already circling Google on an unrelated front when the Commission opened its AI-content investigation: in September 2025, the Commission fined Google 2.95 billion euros for distorting competition in the online advertising technology market -- a separate case, under a separate theory, but evidence that European regulators were treating Google's market power as an active enforcement priority months before AI Overviews became the specific target. APIG's complaint lands inside a regulatory environment already primed to act on Google, not a quiet one.

None of this resolves quickly. France's 2021 fine followed a complaint that had been working through the competition authority for months before it landed; the 2024 fine took even longer. If APIG's complaint follows a similar path, the practical open question is not whether France finds against Google eventually -- its regulator has ruled against Google on this exact underlying issue twice already -- but whether a third finding changes what Google actually does, or becomes a third fine Google treats as a cost of doing business in a market it is not leaving.