According to internal documents obtained from the Wall Street Journal, many people have been calling out Facebook for hate speech since the beginning. This includes Selena Gomez, who is a comedian and only Murders in the Building star.
It all began in 2016, when Gomez visited Menlo Park headquarters of the company to celebrate her account becoming the most followed on Instagram. Gomez received a particularly negative response to her photo opp post. A spokesperson for the Journal said that Gomez reached out to Mark Zuckerberg, COO Sheryl Sandberg, and urged them to act in DMs she shared later.
We now know that Gomez also emailed Facebook executives to express her concerns, thanks to the WSJ reporting. Sandberg replied, saying that Facebook's AI had detected 91 per cent of the 1.5 million posts it took down for violating its hate speech rules. Gomez responded strongly, but that was not enough.
Gomez sent Sandberg an email on October 10, 2020, in which she wrote, "You refuse even to mention, let alone to address, the problem Facebook's with white supremacists, and bigots." According to the Journal, she included screenshots of Facebook pages that she claimed promoted violent ideologies. She also stated that there were many groups "full" of hate and lies that could lead to people being hurt, or worse, even killed.
Gomez apparently emailed her concerns after her private DM sent to Sandberg and Zuckerberg was not answered. Her attempt to shame them publicly for their inaction went nowhere. In a previously shared DM, she stated that Facebook was a "serious issue" and that it was being used to spread hateful misinformation, racism, bigotry, and other inhumane behaviors.
She wrote, "I am calling you both to STOP THIS," in her note. Later, it was made public in an Instagram Story. The Journal's Facebook Files continues to bombshell her email outreach. This is what's new.
In a Journal story, the Journal released documents that revealed Gomez's attempts to contact Facebook leadership. This story demonstrates how Facebook uses AI to detect hate speeches even though its AI is pretty awful. The Journal reports that the platform made a shift in how it used AI to detect hate speech two years ago, and the human reviewers were less focused on this issue. However, the AI of the company has difficulty identifying differences between videos such as first-person shootings and car crashes.
