Yet again this week Silicon Valley PR people are being roiled by news that some of their un-monitored algorithms have allowed for the spread of hate, and in this case, overt anti-Semitism. The situation was at first called out at Facebook, where Pro Publica, acting on a tip, went to purchase ads on Facebook and found that categories like "Jew hater" populated themselves in the ad-targeting system. Reporters there were able to pay Facebook $30 for three promoted posts targeting groups of users who matched the "fields of study" listed as "Jew hater,” “How to burn jews,” or, "Hitler did nothing wrong."
As you can see in the screenshot below, such targets, which also include "NaziParty" as a "field of study", could garner an advertiser some 2,400 potential users.
Facebook immediately responded with one of their rote statements, as Tech Crunch reports, about not allowing hate speech and "We know we have more work to do." And the offending categories were removed, with Facebook's product management director Rob Leathern telling Pro Publica, "we’re also building new guardrails in our product and review processes to prevent other issues like this from happening in the future."
Facebook further explained in this blog post how these categories came to be: "As people fill in their education or employer on their profile, we have found a small percentage of people who have entered offensive responses, in violation of our policies." The company adds, "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."
Joshua Benton of Harvard's Nieman Journalism Lab took to Twitter to point out that you could search for and find some of the same target categories in Google AdWords, with "How to burn jews" and "Jew hater" being the best bets "History of why jews ruin the world" had a "lower search volume, according to the system.
Google further suggested a number of other similar search terms to try such as "burning jews" and "do jews hate christians."
And Google is happy to suggest some other search terms I might want to target to broaden my anti-Semitic reach pic.twitter.com/qZrT4UKigF— Joshua Benton (@jbenton) September 15, 2017
TechCrunch's Devin Coldewey says that perhaps Google should be more quickly forgiven since "Google’s is a more senseless association of queries it’s had and it has less leeway to remove things, since it can’t very well not allow people to search for ethnic slurs or the like." Whereas with Facebook, those target categories came up and appear "quasi-approved (i.e. hasn’t been flagged)" and culled from existing "profile data it thinks fits what you’re looking for."
All this comes on the heels of news this week that Facebook had taken in about $100,000 in advertising money from shadowy Russian sources that posted and promoted fake news stories intent on swaying our November election.
This story is to be continued as we learn all the other deplorable sources from which Facebook has profited without knowing, or at least acknowledging it.
Update: And on Friday, The Daily Beast took a similar look into how Twitter allows advertisers to target users, and they found that an advertiser could choose to market to "26.3 million users interested in the derogatory term “wetback,” 18.6 million accounts that are likely to engage with the word “Nazi,” and 14.5 million users who might be drawn to "n**ger."