Court Orders Google Autocomplete Changes: Japanese Man Defamed by Algorithm

SMS Text
Court Orders Google Autocomplete Changes: Japanese Man Defamed by Algorithm

google autocomplete japanese court order defamation

In a recent cyber-defamation case, a Japanese court ordered Google to suspend autocomplete searches queries related to a specific man’s name. The plaintiff has argued that the autocomplete results, which include suggestions related to crimes he did not commit, have caused him to lose a job and prevented him from finding gainful employment.

Hiroyuki Tomita, the plaintiff’s attorney, is opposed to the autocomplete system and believes that the whole autocomplete system needs to be revised:

“It could lead to irretrievable damage such as a loss of job or bankruptcy just by showing search results that constitute defamation or a violation of the privacy of an individual person or small and medium-sized companies.”

Google has stated they are “reviewing the order,” but the search engine has not complied with the court’s instructions at this time. Today, Google released the following statement:

“A Japanese court issued a provisional order requesting Google to delete specific terms from autocomplete. The judge did not require Google to completely suspend the autocomplete function.”

The autocomplete results are mechanically generated by an algorithm that is based on several factors related to the popularity and volume of search queries.  Since the autocomplete functionality is based on an algorithm and not manually edited, Google continues to argue that the system is not an invasion of privacy:

“Google does not determine these terms manually – all of the queries shown in autocomplete have been typed previously by other Google users.”

This case is not the first time Google has faced legal actions for the autocomplete functionality. After the algorithm came under scrutiny in 2010, Google agreed to alter the autocomplete results and prevent terms from appearing if they were related to illegal piracy. In addition to the piracy case, there is an ongoing case between Google and a UK man who is attempting to remove certain search suggestions related to private stories about his sex life. While the autocomplete feature is mechanically engineered, the algorithm already censors results that are related to registered trademarks, porn, offensive language, piracy, and violence.

Do you think Google should be held responsible for autocomplete searches related to an individual’s name or that the feature improves the search engine?

[Sources Include: BBC News  & The Register]

David Angotti

David Angotti

After successfully founding and exiting an educational startup in 2009, I began helping companies with business development, search engine marketing (SEM), search engine optimization (SEO), conversion rate optimization (CRO), online marketing, mergers and acquisition, product development, and branding. Now, I am focused on a new startup in the travel and tourism market niche.
David Angotti
Get the latest news from Search Engine Journal!
We value your privacy! See our policy here.
  • EDM Machines

    Based on the way Google searches are suppose to work, it would be the fault of the users not the programmers. Autocomplete finishes searches based on others’ search queries. I see no fault with Google at all.

  • Alan Bleiweiss

    Since Google has shown that they are quite capable of censoring the autocomplete feature when it’s in someone’s personal or business interest (copyright, pirating), they have the means necessary to do so in other situations. When a court finds an individual is wrongly harmed through defamatory claims, Google should not hesitate or balk. I don’t care if it’s people’s searches that cause the information to appear. The simple fact that it DOES appear in a Google environment means that Google is thus contributing to the continuation of those defamatory statements, whether intentionally or not. Intention is irrelevant when due process legal action concludes something is defamatory and orders it be taken down.