NEW YORK: Facebook, the social network, has updated its targeted advertising service after learning about the existence of various offensive and extremist categories that had been made available to marketers.

An investigation by ProPublica, the news outlet, found that it could place ads in the Facebook feeds of around 2,300 people who declared an interest in subjects such as “Jew hater”, “How to burn Jews” or “History of ‘Why Jews ruin the world.’”

Facebook reported that these ad categories had been created automatically by its targeted advertising system after a sufficient number of users input certain related terms into their profile.

More specifically, the Menlo Park, California-based enterprise stated in a blog post that a “small percentage” of its members had entered these “offensive” entries in the fields concerned.

“ProPublica surfaced that these offensive education and employer fields were showing up in our ads interface as targetable audiences for campaigns. We immediately removed them,” Facebook’s statement continued.

As the number of individuals making up these segments was “incredibly low”, the social networking giant – which has over two billion active daily users – asserted that the amount of people who could ultimately be targeted was “extremely small”.

In ensuring its platform is a “safe place” for consumers and brands alike, however, the organisation has made some changes to its advertising system. “Keeping our community safe is critical to our mission,” it said.

“And to help ensure that targeting is not used for discriminatory purposes, we are removing these self-reported targeting fields until we have the right processes in place to help prevent this issue.”

One additional step taken by the company involved encouraging advertisers that spot any “inappropriate targeting fields” to flag them up within its ads interface or through the organisation’s Help Center.

Earlier this month, Facebook revealed the results of an internal investigation into potential Russian use of its platform to try and influence the US presidential election.

It found that $100,000 in adspend, and 3,000 ads, could be associated with 470 inauthentic accounts and pages that “were affiliated with one another and likely operated out of Russia” between June 2015 and May 2017. Further investigations into this issue are on-going.

Data sourced from ProPublica/Facebook; additional content by WARC staff