By Sarah E. Needleman 

Twitter Inc. said it has updated its policy on hateful conduct to better protect religious groups, the latest effort by the company to police speech and curb abuse on its platform despite the potential risk of offending some users.

Under the modified terms, tweets with language that "dehumanizes on the basis of religion" will be removed when reported by users or identified by the company itself, the social-media giant said Tuesday. The change was made, in part, in response to a user survey conducted last year that drew more than 8,000 comments, Twitter said.

The policy update covers new tweets as well as those previously posted on the platform. When a tweet is reported and deemed in violation of the policy, a user won't be able to access his or her account until the tweet is deleted, Twitter said.

With the update on hateful conduct, Twitter said it is expanding its focus to protect entire religious groups and not just individuals. Twitter is also evaluating how it could apply the policy regarding offensive language to better protect other groups, including people of color and members of the LGBT community.

The move comes as Twitter and other social-media companies face mounting calls by users and activists to stiffen their treatment of content on their platforms that is deemed offensive and bullying. Twitter in particular has been criticized for ignoring anti-Semitic tweets from Nation of Islam leader Louis Farrakhan and other users.

But Twitter could face more backlash from free-speech advocates and others who have criticized social-media companies for going too far in adjudicating user behavior, for example, in suppressing politically conservative voices. In an interview last month on Fox Business Network, Mr. Trump said, without providing evidence, that Twitter is making it harder for users to follow him on the platform.

A Canadian writer sued Twitter in February, claiming she was unfairly banned because her criticism of transgender rights didn't line up with the company's politics. The case, which was filed in state Superior Court in San Francisco, was dismissed last month.

Twitter last month revealed plans to begin flagging tweets by government officials and political figures who violate its rules about content. Such tweets are now identified as noncompliant and users need to click through a prompt to view them.

On Monday Facebook Inc.-owned Instagram said it was rolling out new features to combat unacceptable behavior, including a tool that relies on artificial intelligence to determine whether language in a post could be deemed offensive and that can prompt the poster to reconsider. Facebook in May took one of its biggest shots yet at online provocateurs, banning an array of personalities whose views it considered too inflammatory including Mr. Farrakhan and talk-show host Alex Jones.

Last month Reddit Inc. quarantined a pro-Trump user group on its platform after it said members violated its policies by threatening violence against law-enforcement officers and public officials.

Social-media companies have also been grappling with increased scrutiny in Washington of their operations and the influence they wield. The Justice Department is currently preparing an antitrust probe into Alphabet Inc.'s Google, while Facebook awaits a possible Federal Trade Commission fine of up to $5 billion over privacy violations.

The White House is planning to hold a Social Media Summit on Thursday to "bring together digital leaders for a robust conversation on the opportunities and challenges of today's online environment." A spokeswoman for Twitter declined to comment on whether the company will participate in it.

Write to Sarah E. Needleman at sarah.needleman@wsj.com

 

(END) Dow Jones Newswires

July 09, 2019 16:53 ET (20:53 GMT)

Copyright (c) 2019 Dow Jones & Company, Inc.
Twitter (NYSE:TWTR)
Gráfica de Acción Histórica
De Feb 2024 a Mar 2024 Haga Click aquí para más Gráficas Twitter.
Twitter (NYSE:TWTR)
Gráfica de Acción Histórica
De Mar 2023 a Mar 2024 Haga Click aquí para más Gráficas Twitter.