Twitter Blocks Ads With Words ‘Illegal Alien,’ Then Reverses Decision as ‘Error’

September 13, 2018 Updated: October 5, 2018

A nonprofit organization was unable to promote several tweets that included the terms “illegal alien” and “criminal alien” on Twitter after the social-media platform invoked its “Hateful Content” policy. Twitter eventually reversed the decision, saying it was “made in error,” after the issue received significant media attention.

The Center for Immigration Studies (CIS), which advocates a “low-immigration, pro-immigrant” policy, tried to add tweets to the center’s Twitter Ads paid promotion campaign, which allows users to promote tweets beyond their follower base. Some tweets were approved, but Twitter refused to allow four of them because of the content.

“A new video from the @DailyCaller showing illegal aliens pouring across the border reminds us why we need a wall,” one of the rejected tweets reads. “Technologies and adequate manpower are well and good, but the best defense is always to prevent individuals from entering in the first place.” The other three tweets included the legal terms “illegal alien” or “criminal aliens” and discussed law enforcement measures against illegal immigration.

The CIS reached out to Twitter and received what appeared to be a standardized reply:

“We’ve reviewed your tweets and confirmed that it is ineligible to participate in the Twitter Ads program at this time, based on our Hateful Content policy. Violating content includes, but is not limited to, that which is hate speech or advocacy against a protected group.”

The CIS documented the issue in a Sept. 12 press release, asking, “Do these tweets illustrate ‘hateful content’, or is Twitter filtering content with a political bias?”

Multiple right-leaning media outlets picked up the story and CIS Executive Director Mark Krikorian was interviewed by Fox News’ Tucker Carlson that night.

During that show, Krikorian learned that Twitter had reversed the decision to block the tweets.

A Twitter spokesperson made a similar statement to The Epoch Times in a Sept. 13 email, saying, “this decision was overturned by our team and was made in error.”

“We enforce our rules dispassionately and judiciously but sometimes mistakes happen—both on the conservative and the liberal side,” the spokesperson said in an email, mirroring the congressional testimonies of the platform’s CEO Jack Dorsey on Sept. 5.

The CIS hadn’t been informed about the unblocking until Sept. 13 afternoon, when it received an email from Twitter that one of the rejected campaigns was accepted. The other three blocked campaigns appeared to have been accepted too, Marguerite Telford, CIS director of communications, told The Epoch Times in a telephone interview.

A Pattern of Bias

Twitter has repeatedly been caught repressing conservative voices.

For several years, Twitter users have accused the company of shadowbanning—hiding a user’s content from other users without informing them.

One method of shadowbanning is the so-called “quality filter,” which removes affected accounts from the “latest” category of search results—unless the user manually switches the filter off. The filter then snaps back on after each search.

The Epoch Times previously reviewed dozens of Twitter accounts of Trump supporters and opponents that appeared to exhibit similar patterns of behavior. Only the Trump supporters were affected by the filter.

The Epoch Times wasn’t able to find any official announcement of this particular function.

Twitter appeared to somewhat relax the filter around July 27 and then rolled it back around the time approaching Dorsey’s announcement on Aug. 31 that he had agreed to testify to Congress.

Also, several Republican congressmen had their Twitter accounts scrapped from the platform’s search suggestion function. When questioned by some media in July, Twitter corrected the situation and said it was the result of a function that improperly penalized accounts for the behavior of their followers. The function affected 600,000 accounts, both liberal and conservative, the company said. But, despite extensive testing, media uncovered only four affected lawmakers—all of them Republicans.

Twitter employees have previously told undercover reporters that the platform had unwritten rules and a company culture that at least condoned suppressing conservative voices.

Dorsey acknowledged in one of his testimonies that algorithms that Twitter uses to filter content may be made biased unintentionally. He said the company is “very, very early” in its work on addressing this issue.

Update: The article has been updated with additional information provided by Marguerite Telford, CIS director of communications.

Follow Petr on Twitter: @petrsvab