German court holds Google liable for AI-generated false answers

The Landgericht München I ruled that Google must be directly liable for inaccurate content produced by its AI‑driven “AI Overviews” feature. The judgment arose after two Munich publishing houses were falsely linked to fraud in AI‑generated search summaries. The court found that these summaries constitute independent statements, not mere extracts of existing web content, and therefore Google cannot invoke the traditional search‑engine liability exemption. Google’s defence—that users are warned to verify sources—was rejected. The decision establishes that the provider of generative AI must ensure the correctness of its outputs and may be held accountable for defamatory or misleading statements. Legal experts note that the ruling could set a precedent for other AI models such as ChatGPT or Claude, prompting broader regulatory scrutiny of AI‑generated information.