Affiliate links on Android Authority may earn us a commission. Learn more.
Court actually holds Google responsible for everything AI Overviews get wrong
- A German court has ruled that Google is ultimately liable when AI Overviews includes errors.
- Publishers found AI Overviews calling their businesses scams, despite a lack of corroborating third party sources.
- The court said that AI Overviews represent original statements made by Google.
Anyone who has used Google’s AI-powered Search tools for long enough already knows: Sooner or later, the AI is going to get something wrong. Maybe that’s just a small oversight, maybe it’s a full-blown hallucination, but it’s impossible to deny that mistakes keep happening. But now Google just might start feeling the pressure to sort some of those issues out, as courts begin ruling that the company is ultimately responsible for any AI-fueled misinformation.
Don’t want to miss the best from Android Authority?
- Set us as a favorite source in Google Discover to never miss our latest exclusive reports, expert analysis, and much more.
- You can also set us as a preferred source in Google Search by clicking the button below.
This particular case arose in Germany, where The Decoder reports that a pair of publishers in Munich discovered that AI Overviews were telling searchers that they engaged in “dubious business practices” and accused them of running scams. After Google didn’t respond to a cease-and-desist letter, they sued.
While Google might get a pass if it were simply surfacing Search results where a third party accused the publishers of fraud, the courts found that AI Overviews represent original statements from Google itself — and as such, the company is liable for their content.
Predictably, Google argued before the court that users have the power to verify the veracity of AI statements by chasing down source links. The court was having none of this, however, rejecting the idea that just because a user could do such legwork for themselves that this should be an excuse to absolve Google of liability, and noting that AI Overviews represent “a self-contained statement with independently understandable content and no reference to other possible interpretations or even unreliable content.”
If that sounds bad for Google, you’re not wrong, and if this trend of accountability continues, we may ultimately see it forced to drastically rethink what kind of information AI Overviews present, lest the company put itself on the hook for anything potentially libelous.
We’ve reached out to Google to see what it thinks about the court’s decision, and will update this post if it gets back to us with any statement.
Thank you for being part of our community. Read our Comment Policy before posting.