New Delhi: The failure of Facebook to moderate content helped fuel religious strife in India and other countries, a report by New York University’s Stern Center for Business and Human Rights has said.
The report cites examples of problematic content shared on Facebook in India – especially targeting Muslims, some by “affiliates” of the ruling BJP – and pointed out that the social media giant failed to take it down.
The observations were part of a broader report where the Stern Center, which aims to study how businesses approach human rights issues in the countries where they function, sought to highlight the importance of content moderation.
The report was released on 8 June and is titled, “Who Moderates the Social Media Giants? A Call to End Outsourcing”. It is meant to trigger better pay and work conditions for content moderators, whom it describes as crucial to keeping internet a safe space.
The report looks at how social media firms Facebook, Twitter, and YouTube moderate their content and the ways the process can be improved.
The moderation shortcomings are a bigger threat with Facebook because it’s the biggest social media network in the world and counts India as its largest market with 300 million users.
Also Read: Zuckerberg uses Kapil Mishra’s ‘Delhi riots threat’ to outline Facebook’s hate speech policy
Rohingyas, Assam Muslims cited as examples
According to a press note accompanying the report, “every day, three million Facebook posts are flagged for review by 15,000 moderators”. Facebook, the press note adds, “has admitted to at least a 10 percent error rate in making content decisions, which suggests 300,000 mistakes a day”.
In the section dedicated to the importance of content moderation in “volatile countries”, the report takes up examples of unmoderated content in India. It starts with the Rohingya refugees from Myanmar who, according to the report, have become “the target of Hindu nationalist hatred”.
The “antagonists, some of them affiliated with Prime Minister Narendra Modi’s Bharatia Janata Party (sic)” have “once again… exploited Facebook in one component of a broader anti-Muslim movement in India”.
As an example, the report cites a Facebook video from 2019 where a “group of men affiliated with the militant wing of the BJP (sic) brandished knives and burned the effigy of a child while screaming, “Rohingyas, go back!”
The report adds that Facebook didn’t take down the post, “on the theory that it was posted by groups claiming to be news organisations and wasn’t directly linked to violence”. “The link may not have been direct, but in June 2019, dozens of Rohingya homes were burned in Jammu, where the video and others like it were shot,” it adds, offering a hyperlink to a report in The New York Times.
Another such incident was the discovery made by the advocacy group Avaaz about “a menacing social media campaign against Bengali Muslims”. The researchers from Avaaz had discovered the social media campaign was using words like “parasites”, “rats”, and “rapists” to describe Bengali Muslims.
In a report published in October 2019, Avaaz said it had directly flagged to Facebook 213 posts of “the clearest examples of hate speech”. However, Facebook only removed 96 pieces of content.
The incident, the NYU report states, also brought to the fore the lack of language expertise in Facebook’s team. The “disturbing posts… were easily found by native Assamese speakers, and yet Facebook’s own team had not previously detected any of them before being alerted to them by Avaaz”.
The NYU report quotes Facebook spokesperson Ruchika Budhraja as saying that the company is paying more attention to content moderation in India and other at-risk countries.
In response to challenges such as these, Facebook is paying more attention to content moderation in India and other at-risk countries, she has been quoted as saying via email.
Meanwhile, the BJP dismissed the NYU report.
BJP national secretary R.P. Singh said reports from “New York Times, Washington Post” cannot be considered evidence of the fact that BJP affiliates use Facebook to incite an anti-Muslim movement. Reports such as these [NYU report] should “cull out data, otherwise it is falsehood and pure propaganda”, he added.
Khemchand Sharma, member of the BJP’s national IT & social media campaign committee, said the report is “nonsense and… trying to defame nationalists across the world”.
“Leftists are paying media houses and others to write anti-national propaganda and trying to incite a level of violence in India similar to what is happening in the US and Europe (Black Lives Matter movement),” he added.
Also Read: Should Twitter, FB censor world leaders like Trump or remain platforms for freedom of speech?
The NYU report seems quite similar to the “research paper” published in The Lancet claiming to provide incontrovertible proof against the drug HCQ for treatment of Covid patients. In its eagerness to run down Mr. Trump it ended up losing its reputation.
Deep down somewhere, we all know who authors such “reports” and what exactly is the purpose behind doing so. This news article quite cleverly hides the identity of the authors who wrote this report. Needless to say, a bit of digging into their profiles will in all possibility yield evidence for the claims made by Mr. Singh and Mr. Sharma.