Two German regional courts have classified Google AI search answers as the company's own statements, making Google directly liable for false claims. The Munich ruling from late May covered AI Overviews that linked two publishers to fraud schemes, while the Aachen order of September 24, 2026 covered AI Mode claims against a physician. Violation of the Aachen injunction carries fines of up to 250,000 euros per case. The development matters because it removes the defense of a neutral intermediary for AI search in Germany.
Munich settlement and Aachen injunction
Google publicly announced an appeal against the Munich Regional Court I decision but did not pursue it to a judgment. According to the source, the dispute ended in a settlement, with no higher court reviewing the reasoning. Whether money changed hands remains unknown, and neither Google nor the affected publisher commented. The Munich court itself declined to assess broader significance beyond that case. As a result, no binding precedent from a higher instance exists so far.
The second case arose after a colleague alerted an Aachen physician on August 28 that AI Mode tied his departure to an alleged sexual assault against his secretary. The doctor had served as chief physician and professor of neuroradiology from 2010 until late June 2026. Follow-up queries with terms such as "suspension," "termination," and "court proceedings" produced detailed multi-paragraph accounts and an invented legal aftermath. The court verified from its own records that the criminal case concerned a different former university professor. The system had transferred another person's history onto the complainant.
The Aachen Regional Court prohibited further distribution of three disputed text variants and acted on an emergency basis without first hearing Google, citing serious risk to professional reputation. The doctor had described the error by email in late August and requested removal, but Google said in mid-September that it would take no action. A formal demand sent the same day also left the texts online, according to the court findings. The order remains challengeable and not final. Each breach threatens a fine of up to 250,000 euros.
What AI liability means for business
For companies deploying AI search, assistants, or summaries, the rulings shift responsibility from data sources to the operator generating the answer. Aachen expressly followed Munich, stating that Google produces independent, new, and substantive statements by evaluating and combining third-party websites. The court added that output could be checked against source sites, and that confronting the model with the doctor's objections would have exposed the errors. Small firms face complaint and injunction costs from a single error, while large operators face hard-to-quantify exposure across millions of generated answers.
Several limits and risks need verification before drawing wider conclusions. Both decisions come from regional courts, the Aachen injunction is not final, and no appellate court has confirmed the publisher-liability theory. The source notes a possible pattern of targeted queries that push models into defamatory combinations, followed by injunctions and settlements. Buyers should ask vendors how outputs are logged, how source comparison is done, how takedown requests are processed, and who bears legal costs. This news alone does not establish copyright claims by publishers.
The marker to watch is whether Google or another defendant takes a case through appeal to a higher court instead of settling. A higher-instance ruling on own-content status would clarify injunction practice and signal whether publishers can seek compensation for source texts used in answers. Continued settlements without judgments would preserve uncertainty and keep takedown speed as the main operational defense.
