Twitter apologizes after research has shown that users can have microtarget ads for Nazis and homophobes
Twitter apologizes after research discovers ad-targeting tool accepts keywords words like “anti-gay” and “white supremacy” for reaching Nazi and homophobic users
- The platform apologizes that users can hate advertisements on target groups
- A BBC report showed that keywords like ‘neo-Nazi’ can be used to reach target groups
- Other keywords were “gay” and “anorexia” or “bulimia”
- Twitter says it has changed its advertising targeting interface
Twitter has apologized after research has shown that the ad targeting interface allows users to reach Nazis and other hate groups.
The possibility of responding to hate groups through targeted advertisements was discovered by a study of the BBC which showed that the Twitter advertising platform enabled users to reach groups by using keywords such as ‘white supremacists’ or ‘transphobic’ and ‘anti-gay’.
“Preventive measures include banning certain sensitive or discriminatory conditions, which we are constantly updating … In this case, some of these conditions were permitted for targeting purposes. This was a mistake, “Twitter said in a statement.
“We are very sorry that this happened and as soon as we were aware of the problem, we solved it.”
Twitter’s ad targeting interface allows users to reach Nazis and other hate groups using keywords such as “white supremacist” or “anti-gay” according to the BBC
In a test, BBC reports that it could use keywords like “neo-Nazis” to actually reach Twitter users in an advertising campaign.
A test where a generic advertisement was conducted in which users were wishing a happy New Year went live for a few hours and reached 37 users before they were removed, the BBC says.
The report also said it was able to target between 67,000 and 81,000 UK users associated with the keyword.
Twitter’s ad-targeting tool was also documented because it could reach sensitive groups such as users with eating disorders such as anorexia or bulimia and was even able to limit the age category of 13 to 24 years old for those users.
An activist interviewed by the BBC says he believes that the ability to target sensitive groups such as people with eating disorders has been abused by companies looking for nutritional supplements or other dubious products.
“I have been talking about my eating disorder on social media for a few years now and have often focused on advertising based on nutritional supplements, weight loss supplements, spinal cord correction,” Daniel Magson, president of the Anorexia and Bulimia Care group told the BBC.
The fear of being able to reach Nazis or other hate groups focuses on the possibility of using targeted advertisements as a tool for recruiting new people into racist organizations or promoting a hateful message.
Twitter CEO Jack Dorsey (pictured above) has come under fire for the platform’s role in promoting hate speech in the past year and has been working on rolling out new policies and tools to reduce toxic content
WHAT IS THE NEW TWITTER POLICY ON DEHUMANIZATION OF CONTENT?
A new policy will act against hate speech aimed at specific religious groups on Twitter.
The policy task for users with flagging content that they think is hateful, at which point the message will be reviewed by the Twitter moderators.
Tweets that predate the policy change will also be deleted, although users will not be suspended.
The platform will continue to assess how and when the new rules should be applied and said it will provide moderators with longer and more robust training.
Twitter has come under fire in the past two years for its shortcomings in reducing hate speech on its platform and has been working on changing policies and enhancing detection efforts.
Last year it banned tweets that were determined to target specific religious groups, especially if they “dehumanize others on the basis of religion.”
This protection for religious groups followed the decision to remove hate speech from political figures.
Messages from political figures that violate Twitter policies are now flagged by some kind of public consent notice that must be read and clicked through before users can access the underlying tweet.
The new policy also limits the reach of the flagged tweets, making it less likely to be seen by a large number of users.