Google offers antisemitic and racist ad targeting too

0
43

Image: Smith Collection/Gado/Getty Images

Facebook isn’t the only advertising giant enabling ad targeting around hateful terms and phrases.


Clickbank Marketing Tools

Google, the world’s biggest advertising company, lets ad buyers target messages around antisemitic and racist search terms like “Jewish parasite” and “Black people ruin everything,” BuzzFeed discovered  this week.

The revelations come as Facebook and Google have had to reckon with the role their huge platforms play in spreading misinformation and hate in the wake of a presidential election in which both ran rampant online.

Google and Facebook collectively control around 70 percent of all online ads, and much of the processes that control them are handled by automation. That’s led to numerous headaches for each as they try to strike a balance between reigning in their platforms’ uglier tendencies and stifling the free flow of information.

Worse yet, Google’s ad platform will offer up an algorithmically generated list of other offensive suggestions when such terms are entered. 

The report comes on the heels of a ProPublica investigation in which the news site bought and received approval for Facebook ads aimed at users who had self-reported interest in topics like “Jew lovers,” “How to burn jews,” or, “History of ‘why jews ruin the world.’”

Facebook cracked down on the problem hours later by removing targeting capabilities around some user-reported fields, including employer, education, and job title. The company blamed its automated systems for having generated the categories based on what a relatively small number of users had put in their profiles.

Unlike Facebook, Google’s options are derived from the billions of searches that happen on its site every day and appear when someone searches the specific term, as opposed to  automatically targeting individuals.

It’s obviously close to impossible for Google to police every single search term that’s funneled into its self-serve ad platform, but there are ways to better filter them with machine learning, as Google itself has demonstrated in the past. 

Google’s SVP of Ads, Sridhar Ramaswamy, claimed in a statement that these particular offensive phrases slipped through the cracks of its enforcement system.

“Our goal is to prevent our keyword suggestions tool from making offensive suggestions, and to stop any offensive ads appearing,” Ramaswamy said. “We have language that informs advertisers when their ads are offensive and therefore rejected. In this instance, ads didn’t run against the vast majority of these keywords, but we didn’t catch all these offensive suggestions.”

“That’s not good enough and we’re not making excuses.  We’ve already turned off these suggestions, and any ads that made it through, and will work harder to stop this from happening again,” the statement continued.

BuzzFeed reported that every term it had used in its test campaign had later been scrubbed save for “blacks destroy everything.”

The ads, which also targeted searches on “the evil jew,” “jewish control of banks,” and “why do Jews ruin everything,” reportedly garnered 17 impressions before they were shut down.

Google previously faced a massive advertiser boycott earlier this year when it was discovered that ads were appearing on YouTube videos from Nazis, terrorists, and other hate groups. The search giant responded forcefully to that controversy with more vetting staffers and automated filtering, but some YouTube creators claimed enforcement became uneven and unaccountable.

This story has been updated with a statement from Google.

Read more: http://mashable.com/2017/09/15/google-hate-terms-ad-buying/

Please follow and like us: