Following Elon Musk’s recent takeover of Twitter, the billionaire quickly laid off about 50 percent of the company’s staff, including members of human rights, safety and integrity teams, and thousands of outside contractors who moderate content in countries around the world, according to the watchdog group Free Press. Executives in charge of privacy and security resigned on November 10, and roughly 1,000 remaining employees followed.
Author and activist Thenmozhi Soundararajan wonders what human rights advocates would do if calls to violence erupt via Twitter in India, where Prime Minister Narendra Modi’s Hindu nationalist movement is stoking religious tension for political gain. “This is not again about hurt feelings for ‘fragile snowflakes,’ this is about direct calls to violence, and an American company is doing that,” Soundararajan told reporters on Monday. “I would assume there is no brand that wants to be linked to genocide.”
Protests erupted as India passed the Citizenship Amendment Act in December 2019, effectively denying full citizenship to Muslims across the country. Twitter struggled to prevent agitators from using the website during Delhi’s deadly riots in 2020, when observers accused India’s police state of complicity in what amounted to an anti-Muslim pogrom. Until recently, human rights groups and Indian civil society were working with Twitter moderators to update the site’s “slur list” after an existing list failed to adequately identify hate speech in the nation’s multiple languages and dialects. This years-long effort to prevent genocidal hate speech in India wasn’t perfect, but at least there was some layer of protection and Indian civil society was involved, according to Soundararajan. Now, former Twitter moderators who worked with civil society are no longer answering emails. Must they tag @ElonMusk in a tweet and wait?
Soundararajan, along with a global coalition of human rights groups and political dissidents, are calling on advertisers to “pause” ad buys on Twitter until Musk can “verifiably” show that his platform will not devolve into a cesspool of violent hate speech, or a tool for tyrants who would spy on their citizens and stifle dissent. Advertising is Twitter’s main revenue stream, and as of last week, at least 50 of the company’s top 100 advertisers had pulled ads from the platform. Companies are often “quiet quitting” to avoid being trolled by Musk’s followers, according to Free Press CEO Jessica J. González. Only a handful of firms issued public statements after dropping ads.
“Even before Musk took over, [Twitter] was dangerous for users, especially outside the U.S., where little moderation occurs,” González said in a press call on Monday, adding that Musk’s chaotic new policies are empowering hate groups and authoritarian politicians. “The U.S. has seen a sharp increase in hate speech, and that is in English. Imagine what users speaking other languages are experiencing.”
Facing global alarm and outrage, Musk recently tweeted a vague graph purporting to show that hate speech “impressions” on Twitter had dropped to “normal” levels after a spike, but the advocates who have worked with Twitter moderators across the world for years have every reason to be skeptical. They said the graph does not state whether the data on hate speech “impressions” was gathered in the U.S. or across the platform globally, and currently Twitter does not have the capacity to accurately measure hate speech — and crucially, the political and ethnic sentiments that fuel violence — on any sort of international level. Regulators in the European Union and countries such as Germany are watching closely. How did Musk compile the data now that a majority of his employees are gone? Independent researchers have documented a steep rise in hate speech on Twitter.
The mainstream media often focus on hate speech on Twitter emanating from the United States, where elections have been marred by misinformation and far right extremists conduct targeted harassment and mass killings, but violence and hate is a complex problem in dozens of countries representing the global majority. White nationalist groups promoting the racist “great replacement” conspiracy theory have used Twitter to network globally, for example. Twitter is also a crucial communications service for nations across Africa and Asia, where content moderation was already sparse and reliant on outside assistance from civil society before Musk sparked a mass exodus of contractors and employees.
“Facebook is harmful, but in our markets, they are doing a better job at listening and trying to work with people on the ground to institute these protective measures,” said Rosemary Ajayi, a lead researcher at Digital Africa Research Lab, who worked with Twitter’s team in Nigeria. Ajayi tracked content in Nigeria and other countries that clearly violated Twitter’s rules and how the company responded when the content was reported, an in many cases it took Twitter up to three months to respond, if moderators responded at all.
“How does that make sense during an election weekend, when you are responding three months later?” Ajayi said.
Working with a skeleton crew and facing massive potential revenue losses, Musk has simultaneously tried to calm advertisers while using his own Twitter account to spread misinformation, taunt critics, troll (or suspend the accounts of) public figures, and conduct “polls” over sweeping changes to the platform’s longstanding community standards. Advocates say Musk’s polls are arbitrary and likely compromised by bots and trolls.
Guided by his polls, Musk is reinstating accounts for users previously booted for violating violence and hate speech rules. Beneficiaries of the new policy include former President Donald Trump, who came under scrutiny for his role in the January 6 riot at the Capitol. Meanwhile, multiple anti-fascist accounts were suspended after Musk took over.
David Duke, a notorious white supremacist, was recently allowed back on Twitter before activists protested, according to Wendy Via, president of the Global Project Against Hate and Extremism. Via said the video taken by the perpetrator of the 2019 mass shooting at a mosque in Christchurch, New Zealand, was also recently removed from the platform after being re-posted by far right accounts, perhaps to test the waters of moderation under Musk.
“The video of the shooting [was] once again freely circulating on Twitter — it was reported and removed, but given how the bad actors are working right now, it’s probably circulating as we speak,” Via told reporters on Monday. “We just got David Duke off; imagine if [Musk] allows all of these people come back?”
Advocates beyond the U.S. say leaving Twitter altogether is simply not an option, especially in lower-income countries where users rely on Twitter to communicate with each other and local officials during disasters and emergencies. Ajayi said most Americans would be shocked by the experience of Twitter users in Nigeria and other African nations. In the U.S., users are complaining about paying $8 a month to be “verified” by Twitter, which declares an account authentic with an iconic blue check. In Nigeria, Ajayi said, users will pay up to $5,000 on the black market for a verified account on Twitter and other platforms — and all the online power and influence it brings.
“I’m African, I am not leaving Twitter,” Ajayi said. “I need to be monitoring the bad actors, so I will be the last one there.”