From Matthias Bastian (The Decoder) Tuesday:
Landmark German ruling declares Google’s AI Overviews are Google’s own words and makes it liable for false answers ….
The Regional Court of Munich hit Google with a temporary injunction barring the company from spreading false claims about two Munich-based publishers through its AI-generated search overviews (case no. 26 O 869/26). The court classified Google as a direct infringer because the “AI overview” is its own content, not just a list of search results.
Google’s AI overviews had falsely tied two publishing companies to scams, subscription traps, and shady business practices for certain search queries. According to the court, the AI mixed up information about other, genuinely sketchy companies with the plaintiffs and drew connections that didn’t appear in any of the linked sources. The publishers sent Google a cease-and-desist letter, but Google didn’t respond appropriately….
Google’s AI overviews work nothing like traditional search results, the court argues. The AI rewrites and judges results “in its own words and according to its own structure,” the ruling says. In the case at hand, for example, it opened with confident claims like “Yes, [company] is known for dubious business practices,” then built its own structure with a summary, red flags for the alleged scam, and tips for users.
The court also found that the AI overview made claims “that are not even made in the search results.” None of the linked sources drew any connection between the plaintiffs and the shady companies the AI mentioned. The court called these “the defendant’s own statements.” …
At the hearing, Google argued that users could check the linked sources themselves to verify whether the AI summary was correct. Users generally knew “that information generated with AI should not be blindly trusted,” the company claimed….
The court rejected this. The possibility of disproving a statement through further research doesn’t “regularly exempt from liability for this statement.” …
I am told the judgment itself is this one, though I haven’t reviewed a translated copy myself.
Under American law, I think Google AI would likewise be potentially subject to liability even though Google Search wouldn’t be: 47 U.S.C. § 230. As I wrote in 2023,
Section 230 states that, “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” “[I]nformation content provider” is defined to cover “any person or entity that is responsible, in whole or in part, for the creation or development of information provided through the Internet or any other interactive computer service.”A lawsuit against an AI company would aim to treat it as publisher or speaker of information provided by itself, as an entity “that is responsible, in whole or in part, for the creation or development of [such] information.”
As the leading early § 230 precedent, Zeran v. AOL, pointed out, in § 230 “Congress made a policy choice … not to deter harmful online speech through the … route of imposing tort liability on companies that serve as intermediaries for other parties’ potentially injurious messages.” But Congress didn’t make the choice to immunize companies that themselves create messages that had never been expressed by third parties. Section 230 thus doesn’t immunize defendants who “materially contribut[e] to [the] alleged unlawfulness” of online content.
An AI company, by making and distributing an AI program that creates false and reputation-damaging accusations out of text that entirely lacks such accusations, is surely “materially contribut[ing] to [the] alleged unlawfulness” of that created material….
Relatedly, traditional § 230 cases at least in theory allow someone—the actual creator of the speech—to be held liable for it (even if in practice the creator may be hard to identify, or outside the jurisdiction, or lack the money to pay damages). Allowing § 230 immunity for libels output by an AI program would completely cut off any recourse for the libeled person, against anyone.
In any event, as noted above, § 230 doesn’t protect entities that “materially contribut[e] to [the] alleged unlawfulness” of online content. And when AI programs output defamatory text that they have themselves assembled, word by word, they are certainly materially contributing to its defamatory nature.
Likewise, as I argued in my Large Libel Models? Liability for AI Output, if an AI company is alerted that its software is regularly outputting specific false statements about a person or business, and doesn’t take prompt and reasonable steps to stop that, it could be held liable even under a knowing/reckless falsehood. (I also argued that it could be liable for proven damages caused to private figures even under a negligent design standard.)
In any event, I can’t speak to whether the German court’s decision is consistent with broader principles of German law, but from the press account it sounds like the analysis under U.S. law would likely be in many ways similar (except for the temporary injunction, which I expect would be harder to get, though not necessarily impossible). For more posts on Large Libel Models, see here.
