Munich’s Regional Court (case 26 O 869/26) issued a temporary injunction against Google’s AI Overviews on May 28, calling AI-generated search summaries Google’s own speech, not indexed third-party content.
Two Munich publishers triggered the case after AI Overviews called them “known for dubious business practices” and linked them to scams and subscription traps. None of those claims appeared in Google’s linked source pages. Google’s AI had mixed up these publishers with genuinely sketchy businesses and invented the connections wholesale.
Google ran its standard playbook: AI outputs aren’t reliable, users know this, therefore we’re not liable. The court didn’t buy it. AI Overviews generate “independent, new, and substantive statements” by synthesizing source material, and only Google can correct those outputs. After publishers sent a cease-and-desist and Google did nothing, the court said Google “must be held accountable.” Google covers 80% of legal costs.
This may be the first ruling holding any AI firm directly liable for AI speech. Every AI product generating summaries, from Perplexity to Bing Copilot, just got a case study for why “we disclaim accuracy” won’t hold up when your model makes affirmative false statements about identifiable businesses. The court called those outputs “primarily an expression of defendant’s commercial activity” — not protected speech.
If you’re shipping an AI search or summarization feature, your legal team needs to read case 26 O 869/26 before Monday. Disclaimer banners don’t cover affirmative false statements.
Nathan Zakhary