Actor and singer Selena Gomez turned out to be angry directly at the boss of Facebook because of the problem of hate speech. This was revealed from Facebook's internal documents on artificial intelligence (AI) and hate speech obtained by the Wall Street Journal.
It all started when Gomez visited Facebook's headquarters in Menlo Park, California in 2016 to celebrate her achievement as the account with the most followers on Instagram.
When she saw the unpleasant response in one of her posts about the visit, Gomez then contacted Facebook founder and CEO Mark Zuckerberg and Facebook COO Sheryl Sandberg ahead of the 2020 US presidential election via Instagram DM.
In his first DM, Gomez asked Zuckerberg and Sandberg to stop the spread of hate speech and misinformation. It turned out that after sending the DM, Gomez also contacted Sandberg and other Facebook executives via email.
Sandberg replied to the email and said that Facebook's AI managed to detect 91% of the 1.5 million posts they removed for violating hate speech rules. But that explanation wasn't enough for Gomez.
"You guys refuse to even mention, let alone discuss, the problems Facebook has with white supremacy and bigots," Gomez said in an email sent on October 10, 2020, as quoted from Mashable, Tuesday (19/10/2021).
Gomez also included screenshots of several Facebook groups she said promote violent ideology. This 29-year-old woman also said there are many Facebook groups full of hate and lies that can harm many people.
Gomez apparently submitted his complaint via email because the DM he previously sent to Zuckerberg and Sandberg was not responded to. Gomez also uploaded the contents of the DM to Instagram Stories, but even after it was made public, Zuckerberg and Sandberg did not reply to the message.
The document exposing Gomez's attempts to contact Facebook officials was released as part of a WSJ report on how Facebook is using AI to detect hate speech on its platform.
According to the WSJ report, for the past two years Facebook has relied on AI to detect hate speech and override the role of human moderators. Yet Facebook's AI had a hard time identifying the difference between videos of shootings, car crashes and cockfighting.
This report was published at a time when Facebook was facing a lot of problems. Recently, whistleblower and former Facebook employee Frances Haugen said Zuckerberg was prioritizing profits and viral posts over user safety.
