From hate speech to nudity, Facebook’s oversight board picks its first cases

By Elizabeth Culliford



graphical user interface: FILE PHOTO: Facebook logos


© Reuters/JOHANNA GERON
FILE PHOTO: Facebook logos

(Reuters) – Facebook Inc’s independent Oversight Board announced on Tuesday the first six cases where it could overrule the social media company’s decisions to remove certain pieces of content from its platforms.

The board, which Facebook created in response to criticism of its handling of problematic content, said it had received 20,000 cases since it opened its doors in October.

Three of the six cases involved content that Facebook removed for breaking hate speech rules.

An Oversight Board spokesman said hate speech cases had been “the most significant proportion” of appeals received.

“Hate speech is an especially difficult area,” Jamal Greene, one of the board’s co-chairs and a professor at Columbia Law School, told Reuters in an interview. “It’s not that easy … for an algorithm to get the context of” such speech.

Gallery: A timeline of voting rights in

Read More

I tried Parler, the social media app where hate speech thrives

While the Bay Area hosts juggernauts Facebook and Twitter, neighboring Nevada is home to social media upstart Parler, headquartered in Henderson, a suburb of Las Vegas. Fear and loathing are apt descriptors for the site, where the rancor of the far-right thrives. Beneath the thin guise of the app’s self-proclaimed emphasis on “free speech” lies the ability to say not just a hypothetical “anything,” but specifically to share racist slurs and violent threats toward political opponents. On Parler, Nazi imagery flourishes, death threats abound, and conspiracy theories reign.

Following Twitter and Facebook’s recent efforts to label false and misleading posts, users flocked to Parler, which has occupied a top 10 spot on both the Android and Apple app stores in recent weeks. Parler users bemoan big tech, but the app’s funding comes at least in part from Rebekah Mercer, daughter of Robert Mercer, a billionaire with a history of data-mining.

Read More

Countering hate on social media

Study: Countering hate on social media
Figure 1 from the paper: Examples of Twitter conversations (reply trees) with labeled hate (red), counter (blue), and neutral speech (white). The root node is shown as a large square. Credit: Garland et al, EMNLP 2020

The rise of online hate speech is a disturbing, growing trend in countries around the world, with serious psychological consequences and the potential to impact, and even contribute to, real-world violence. Citizen-generated counter speech may help discourage hateful online rhetoric, but it has been difficult to quantify and study. Until recently, studies have been limited to small-scale, hand-labeled endeavors.


A new paper published in the proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) offers a framework for studying the dynamics of online hate and counter speech. The paper offers the first large-scale classification of millions of such interactions on Twitter. The authors developed a learning algorithm to assess data

Read More

Facebook releases election and hate speech enforcement data

The company estimated it helped register 4.5 million voters in the United States this year across Facebook, Instagram and Messenger, and helped 100,000 sign up to be poll workers. Since its launch, 140 million people have visited the company’s voting information center, and on Election Day, 33 million people visited its election center, which included results as they came in.

The report comes days after chief executive Mark Zuckerberg was grilled about Facebook’s handling of content during the election on Capitol Hill. He said at the time that Facebook was working on a post-mortem of its election actions but did not say when it might be completed.

The prevalence of hate speech continues to be a problem on Facebook. About one out of every 1,000 things users see on the flagship site contains hate speech, Facebook said in its third-quarter Community Standard Enforcement Report. It did not release a similar

Read More

Facebook Estimates Hate Speech Seen in 1 Out of 1000 Views on Its Platform | World News

(Reuters) – Facebook Inc for the first time on Thursday disclosed numbers on the prevalence of hate speech on its platform, saying that out of every 10,000 content views in the third quarter, 10 to 11 included hate speech.

The world’s largest social media company, under scrutiny over its policing of abuses, particularly around November’s U.S. presidential election, released the estimate in its quarterly content moderation report.

On a call with reporters, Facebook’s head of safety and integrity Guy Rosen said that from March 1 to the Nov. 3 election, the company removed more than 265,000 pieces of content from Facebook and Instagram in the United States for violating its voter interference policies.

Facebook also said it took action on 22.1 million pieces of hate speech content in the third quarter, about 95% of which was proactively identified. It took action on 22.5 million in the previous quarter.

The company

Read More