Court Holds Google Liable for Incorrect AI Answers
In a decision expected to impact other artificial intelligence (AI) developers, a German court determined that Google is responsible for answers generated by its AI features, including for incorrect answers created by the search site’s AI overview.
The case presented the question about whether the AI service should be held to the same legal standard as traditional search results.
Two Munich-based publishing companies filed lawsuits against the company. Google's AI overview had incorrectly linked those companies to questionable business practices and fraudulent schemes, inventing connections that did not exist, according to media outlet Deutsche Welle.
Even after the publishers sent the tech company a cease-and-desist letter earlier in 2026, Google did not correct the misleading information, according to Ars Technica.
Typically, search engines have some protection against allegations of defamation because the search results redirect users to external content that is not controlled or owned by the search engine.
Although the company argued it was not responsible for data processing and that the content in the AI overview was not owned by Google, the court disagreed.
The Bavarian court found that the Google AI overviews did not simply present a list of links, but instead make “independent, new, and substantive statements” based on its interpretation of links generated by a search query, Ars Technica reported. This means that only Google is able to correct the algorithm supporting the feature and generated response from the AI overview. The court also decided that a disclaimer cautioning users to independently verify information provided by the overview was insufficient in protecting the company against liability for incorrect answers.
Ultimately, the court ruled on 12 June that the AI summary is content that can be directly attributed to Google, leaving the company and other AI developers open to future defamation lawsuits. The court also approved a temporary injunction ordering Google to stop spreading false claims in its AI overviews.
Google announced it will appeal the ruling. “We invest deeply in the quality of AI Overviews to ensure that the overwhelming majority of responses provide accurate information, and they are designed to reflect the information that exists on the web. We’re carefully reviewing this decision, which is not yet final,” a Google spokesperson told Ars Technica.
The court also determined that the statements generated by an AI overview or similar feature do not have free speech protections. While a person’s speech and opinions may have those protections, AI statements are the product of an algorithm. (Landgericht München I (District Court Munich I), No. 26 O 869.26, 2026)
Explore more about the legal ramifications and limitations around AI use in security in Security Management's coverage from earlier this year.
