Google Owns What Its AI Answers — Two German Rulings and What They Mean for Your Shop

TL;DR — Two German courts ruled in 2026 that Google's AI Overviews and AI Mode answers are Google's own statements, and Google answers for them. If AI search says something false about your brand, there's now a legal path — but you still have to catch and report it first.

What happened

For years, when Google Search surfaced a wrong link, Google's defense was "we only display links; the content belongs to the site." German courts rejected that defense for AI-generated answers. Two cases matter:

• Munich (June 2026): Two Munich publishers sued after AI Overview linked them to dubious business practices, subscription traps and fraudulent schemes — connections that did not exist. The court ruled that because the AI summarizes and evaluates content "in its own words," it creates new, independent statements attributable to Google, not a mere list of links. Google was ordered to stop spreading the claims and to bear 80% of the legal costs.
• Aachen (September 2026): Google's AI Mode told users that a doctor and professor had been convicted of a sex crime. The conviction actually belonged to a different man. He asked Google to delete it on August 28 and sent a formal legal warning on September 15; neither worked. Per MadRobot's report, the Aachen Regional Court banned Google from repeating the claims without even hearing Google first, citing the urgency of false claims "publicly visible" to an unlimited audience — with fines up to €250,000 per breach. That case settled rather than being tested on appeal.

The Munich court also rejected Google's argument that users "know AI answers shouldn't be trusted blindly" and can check the sources. The court called an AI overview "a self-contained statement with independently comprehensible content" shown without any indicator of unreliability.

Why this matters for your shop

You probably don't have a telecom company's legal weight — but the risk is real: anyone can mention your brand and the AI Overview may weave a mixed-up story around it. Because AI writes answers "in its own words," it sometimes confuses two entities: your company with a similarly named one, an old news story with your brand, or one city with another. Two pieces of good news:

1. This is no longer "just how AI is." Before these rulings, complaining about a wrong AI answer was like complaining about the weather. Now, at least in Germany, there's a clear legal route — and the court explicitly said checking would have been easy.
2. Detecting the false answer is still your job. Google is accountable, but accountability starts with a report. Nobody reports, nobody gets fixed.

The two rulings side by side

| | Munich (June 2026) | Aachen (Sept 2026) | |---|---|---| | Plaintiff | Two Munich publishers | A doctor and professor | | False claim | Linked publishers to fraud and subscription traps | Attributed another man's conviction to him | | Feature | AI Overview | AI Mode | | Outcome | Publication ban + Google pays 80% of costs | Ban without hearing Google + up to €250,000 fine per breach | | Status | Google appealed (not final) | Settled without appeal |

Checklist: if AI search says something false about your brand

• [ ] Screenshot with date and URL — AI answers change daily, so preserve evidence of what was shown
• [ ] Check on multiple devices and browsers: is the answer wrong everywhere or in one variant?
• [ ] File a formal request through Google's official tools (Search Console, removal request forms) — the documented request date was decisive in the Aachen case
• [ ] If the false claim seriously damages the business, consult a lawyer — the German rulings opened a path even though appeals elsewhere are still pending
• [ ] Find the source of the error: it's often one old page or news story the AI mixed up — fixing or removing the source fixes the AI answer long-term
• [ ] Strengthen your own site with clear brand facts (full name, history, contact info) so the AI has less to guess about

Sources

• German courts rule Google liable for what its AI search says — MadRobot
• German court holds Google liable for fake AI answers — DW

Q: The rulings are German — what's in it for businesses elsewhere?

You can't cite them directly outside Germany, but two things changed: first, there's now an authoritative legal document saying AI answers carry direct liability — which strengthens your negotiating position in any complaint, even one filed with Google itself. Second, the same reasoning is spreading to other jurisdictions; the Munich ruling aligned with earlier European cases and likely won't be the last.

Q: How do I find out what Google's AI says about my brand?

Once a week, search your brand name plus common combinations (name + city, name + service, name + "scam" or "lawsuit") and read the AI answer. Screenshot it. Tools like Search Console's AI Mode reports show traffic but not the answer's content — manual checking is still necessary.

Q: Does this mean traditional SEO doesn't matter anymore?

The opposite. AI answers are built from web content; if there's no correct, authoritative content about your brand, the AI fills the gap with wrong or outdated sources. The German rulings raised the cost of being absent: instead of just losing rankings, a false story built about you may acquire legal consequences.