German Court Says Google Can Be Held Accountable for AI Overviews Errors, Company Responds

Google has responded after a court in Germany ruled that the company can be held responsible for inaccurate information generated by its AI Overviews feature. The decision relates to AI-generated summaries that allegedly published false claims about two publishers. The Mountain View-based tech giant said it is reviewing the ruling and noted that the decision is not final. The case is attracting attention because it could shape how courts approach responsibility for errors generated by artificial intelligence systems.

Google Responds as German Court Scrutinises AI Overviews

The company told Android Authority that its AI Overviews are designed to reflect information available on the web and that the vast majority of responses generated by the feature are accurate. The company also pointed to existing measures intended to identify and address misleading or incorrect AI-generated content.

The response follows a ruling by the Regional Court of Munich in a case involving two Munich-based publishers. According to a report by The Next Web, the publishers challenged AI-generated summaries that linked them to scams, subscription traps, and questionable business practices.

The publishers reportedly argued that the claims were inaccurate and did not appear in the sources referenced by the AI summaries. The court reportedly found that the system generated information unsupported by the cited material and incorrectly linked the publishers to unrelated businesses.

A central issue in the case was whether AI Overviews should be treated like traditional search results. The court reportedly concluded that the feature generates its own summaries, rather than simply directing users to third-party websites. Based on that reasoning, the judges considered the disputed statements to be Google's responsibility.

The court is also said to have rejected the argument that users could independently verify the information by reviewing the cited sources. According to the ruling, providing links does not absolve responsibility for inaccurate claims in an AI-generated summary.

The court's reasoning reportedly prevented Google from relying on legal protections commonly available to platforms that host third-party content. The judges reportedly viewed the summaries as content produced by Google's systems rather than information supplied directly by external websites.

Notably, the ruling remains a preliminary decision and does not establish a binding legal precedent. Google can appeal the outcome, and further proceedings could alter the final result. The German court reportedly recently dismissed a separate complaint involving AI Overviews, though it did not rule out liability in similar cases.

According to the report, the Munich court ordered Google to cover 80 percent of the legal costs related to the proceedings. The company had not publicly commented on the ruling before providing its statement to Android Authority.

The case could have implications beyond Google. If higher courts uphold the Munich court's interpretation, similar questions may arise for other AI-powered search-and-answer platforms that generate responses directly for users. The dispute also comes as regulators across Europe continue to examine the legal responsibilities of companies developing and deploying generative AI systems.