Trying to quell an unprecedented ad boycott, Facebook recently agreed to modify its content policies, including the prohibition of a wider category of hateful language in ads, as well as labeling on posts from political figures that violate hate-speech terms of service.
Yet, recent analysis of Facebook’s ad library by the Institute on Strategic Dialogue (ISD) found delays or mistakes in policy enforcement that enabled hateful and harmful content to spread through paid targeted ads.
In short, it doesn’t matter what policies Facebook puts into place if it doesn’t enforce them, according to Aoife Gallagher, an analyst at the London-based think tank.
“The key issue lies in policy enforcement, not in policy design, in these instances,” states Gallagher.
Using a series of keyword searches related to protests that following the murder of George Floyd, ISD turned up several ad campaigns that appeared to violate Facebook’s ad policies.
These included a network of Pages running what appeared to be a scam by targeting older users with protest-related ads promising “free” books in exchange for credit-card information. In exchange for running these ads, Facebook received more than $170,000, by ISD’s estimate.
Another such campaign consisted of ads posted by political representatives and candidates advocating violence toward protesters.
“Despite improvements to both the policies around political advertising on Facebook and the associated transparency enabled by the platform’s ad library, there remain critical issues with the enforcement of policies that ostensibly ban violent, fraudulent or misleading advertising,” Gallagher notes in the new report.
“Even where action is taken on violating ad content, such a response is often reactive and delayed, after hundreds, thousands, or potentially even millions of users have already been served those ads on their feeds,” Gallagher adds.
For example, it recently took Facebook an estimated 24 hours before removing 88 ads bought by President Trump’s re-election campaign, which contained Nazi symbols.
During that time, the ads garnered over 1 million impressions, by ISD’s count.
Added Gallagher: “Vetting processes are supposedly in place for ads before they launch to the public, but many that violate Facebook’s Terms of Service, and which promote violence, false information or scams, still make it through the net and ultimately profit the platform.”Facebook did not return requests for comment by press time.