Facebook disputes claim of inadequate flagging of vernacular content

Facebook allegedly lacks adequate ability to flag questionable content in India’s two most spoken languages --Hindi and Bengali- according to internal company documents cited by a whistleblower. The company has disputed this saying it uses technology to proactively detect hate speech in the two languages as well as have reviewers flagging such content in 20 Indian languages.

The social network, which has been under fire for reportedly prioritising profits over user safety following revelations made by a former employee in an anonymous complaint to the US Securities and Exchange Commission (SEC), allocated just 13% of its budget to counter misinformation and hate speech during the financial year 2020 to markets outside of the United States, including India, according to the internal company documents cited by the whistleblower.

In this interview with the Economic Times, Urvashi highlights Facebook's limited investments in content moderation in the global south.

Researchers noted that the whistle blower’s claims are important also because it reveals that the “issue is not about free speech and content moderation. Rather...the types of algorithms that are used, what they optimise for, what type of content they amplify, is what we need to pay more attention to.”

"Most Facebook’s algorithms clearly optimise for its profitability rather than user safety and well-being," said Urvashi Aneja, an associate fellow at Chatham House and a founding director at Tandem Research.

Read the piece here.