- By Marianna Spring
- Correspondent for BBC disinformation and social media
March 5, 2023
Updated 1 hour ago
picture description,
Elon Musk took control of Twitter in October 2022
Twitter insiders have told the BBC that following layoffs and changes under owner Elon Musk, the company can no longer protect users from trolling, state-coordinated disinformation and child sexual exploitation.
Exclusive academic data and testimony from Twitter users backs up their claims and suggests that hate has thrived under Mr Musk’s leadership, with trolls encouraged, increased harassment and a rise in accounts following misogynistic and abusive profiles.
Current and former employees of the company tell the BBC Panorama that features designed to protect Twitter users from trolling and harassment are proving difficult to maintain amid what they describe as a chaotic work environment where Mr Musk is constantly being followed by bodyguards. I’ve spoken to dozens, some recorded for the first time.
The former head of content design says everyone on her team that created security measures like nudge buttons has been fired. She later resigned. Internal research from Twitter suggests these security measures reduced trolling by 60%. An engineer working for Twitter told me “no one does that kind of work now,” likening the platform to a building that appears fine from the outside but is “on fire” inside.
Twitter has not responded to the BBC’s request for comment.
My research also found:
- Concerns that child sexual exploitation is on the rise on Twitter and is not being adequately raised with law enforcement
- Targeted harassment campaigns aimed at curbing freedom of expression and foreign interference — once a day removed from Twitter — go “undetected,” according to a recent staffer.
- Exclusive data showing how misogynistic online hate targeting me has increased since the acquisition and that there has been a 69% increase in new accounts following misogynistic and abusive profiles.
- Rape survivors have been targeted by accounts that have become more active since the takeover, with evidence suggesting they have been restored or recreated.
Abuse on Twitter is nothing new to me – I’m a reporter sharing my coverage of disinformation, conspiracies and hate. But for most of the past year, I’ve noticed that it’s been steadily declining across all social media sites. And then in November I realized that it had gotten worse again on Twitter.
It turns out I was right. A team from the International Center for Journalists and the University of Sheffield have been tracking the hate I’m receiving and their data showed that the abuse directed at me on Twitter has more than tripled compared to the same since Mr Musk took over Period of the year before.
All social media sites have been under pressure to take action against online hate and harmful content – but they say they are taking action to deal with it. Measures that no longer seem to be high on the agenda on Twitter.
In San Francisco, Twitter’s headquarters, I went in search of answers. Where better to get them than from a technician responsible for the computer code that makes Twitter work. Since he’s still working there, he asked us to hide his identity, so we’ll call him Sam.
“For someone inside, it’s like a building with all the parts on fire,” he revealed.
“From the outside the facade looks good, but I can see that nothing is working. All the plumbing is broken, all the faucets, everything.”
He says the chaos was caused by the major staffing disruption. At least half of Twitter’s workforce has been fired or selected to leave the company since Musk bought it. Now people from other teams have to shift their focus, he says.
“A whole new person, without the expertise, is doing what used to be done by 20+ people,” says Sam. “That leaves room for a lot more risk, a lot more opportunity for things to go wrong.”
He says previous features still exist, but those who designed and maintained them are gone – he believes they are now unmanned.
“There are so many things broken and no one cares that you see this contradictory behavior,” he tells me.
He believes the extent of the confusion is because Mr. Musk does not trust Twitter employees. He describes how he called in engineers from his other company — electric car maker Tesla — and asked them to evaluate the engineers’ code in just a few days before deciding who to fire. It would take “months” to understand such code, he tells me.
He believes that lack of trust is betrayed by the level of security Mr Musk surrounds himself with.
“Wherever he goes in the office, there are at least two bodyguards – very bulky, tall, Hollywood movie -[style] bodyguard. Even if [he goes] to the toilet,” he tells me.
He thinks Mr. Musk is about the money. He says all cleaning and catering staff have been laid off – and that Mr Musk even tried to sell the office plants to the staff.
picture description,
Lisa Jennings Young, former head of content design, says her entire team has been cut
Lisa Jennings Young, former head of content design at Twitter, was one of the people who specialized in rolling out features designed to protect users from hate. Twitter was a hotbed of trolling long before Mr. Musk took over, but she says her team has made good progress in curbing it. Internal Twitter research viewed by the BBC appears to confirm this.
“It wasn’t perfect at all. But we tried and made things better all the time,” she says. It is the first time she has spoken publicly about her experience since leaving after Mr Musk took over.
Ms. Jennings Young’s team worked on several new features, including Safe Mode, which can automatically block abusive accounts. They also designed labels for misleading tweets and what they call a “harmful reply nudge.” The “Nudge” warns users before they send a tweet in which AI technology has detected trigger words or malicious language.
Twitter’s own research, viewed by the BBC, appears to show that the nudge and other security tools are effective.
picture description,
Research shows that abuse targeting me on Twitter has more than tripled since Musk acquired it
“Overall, 60% of users deleted or edited their answer when given the nudge a chance,” she says. “But what was even more interesting is that after we nudged people once, they wrote 11% less harmful responses in the future.”
According to data compiled by the University of Sheffield and the International Center for Journalists, these security features were implemented around the time my abuse on Twitter seemed to be winding down. It’s impossible to directly correlate the two, but given what the evidence tells us about the effectiveness of these measures, it’s possible to make a connection.
But after Mr Musk took over the social media company in late October 2022, Lisa’s entire team was fired, and she herself chose to leave in late November. I asked Ms. Jennings Young what happened to features like the harmful response boost.
“There’s no one around to work on it right now,” she told me. She has no idea what happened to the projects she did.
So we tried an experiment.
She suggested a tweet that she expected would trigger a nudge. “Twitter employees are lazy losers, jump off the Golden Gate Bridge and die.” I shared it on a private profile in response to one of her tweets, but to Ms. Jennings Young’s surprise, no nudge was sent. Another tweet we shared that contained abusive language was picked up – but Lisa says the nudge should have caught a message wishing a user dead, not just swear words. As Sam had predicted, it didn’t seem to be working out as planned.
During this investigation, I received messages from many people telling me how the hate they receive on Twitter has increased since Mr Musk took over – sharing examples of racism, anti-Semitism and misogyny.
Ellie Wilson, who lives in Glasgow, was raped while studying at university and began posting about the experience on social media last summer. She received a supportive response on Twitter at the time.
picture description,
Rape survivor Ellie Wilson says she’s noticed a surge in hateful messages lately
But when she tweeted about her attacker in January after his sentencing, she was met with a spate of hate speech. She received abusive and misogynist responses – some even telling her she deserved to be raped.
“[What] I find it the most difficult [is] the people who say I wasn’t raped or that it didn’t happen and that I’m lying. It’s sort of like secondary trauma,” Ms. Wilson told me.
Their Twitter following was smaller prior to the takeover, but when I looked at accounts that were targeting them with hate this time around, I noticed that the trolls’ profiles had become more active since the takeover, suggesting that they were previously banned and recently had been reinstated.
Some of the accounts had even been set up around the time of Mr. Musk’s takeover. They seemed intent on spreading hate with no profile pictures or identifying marks. Several are following and interacting with content from popular accounts accused of promoting misogyny and hatred – which were restored to Twitter after Musk decided to restore thousands of suspended accounts, including that of controversial influencer Andrew Tate.
“By giving these people a platform, you empower them. And you say, ‘That’s fine, you can do that.’”
Some of the accounts also targeted other rape survivors with whom she is in contact.
Andrew Tate did not respond to the BBC’s request for comment.
New research from the Institute for Strategic Dialogue – a UK think tank that studies disinformation and hate – mirrors what I have uncovered about the troll accounts targeting Ellie.
It shows that since Mr Musk took over, tens of thousands of new accounts have been created, which then immediately followed known abusive and misogynistic profiles – 69% more than before his tenure.
Research suggests these abusive networks are now growing – and that Mr Musk’s acquisition created a “permissive environment” for the creation and use of these types of accounts.
Elon Musk’s Twitter storm
Panorama examines how Elon Musk’s ownership is transforming one of the world’s most influential social media platforms.
Available to watch on BBC One at 20:00 GMT, Monday 6 March
Musk’s top priorities since the acquisition, according to his tweets, are making the social media company profitable and championing freedom of expression.
In December 2022, he released internal documents called “Twitter Files” to explain why he felt the company hadn’t applied its moderation and suspension policies fairly under the old leadership.
But those who have been inside feel that Mr Musk has used this to de-priority overall user protection. Even the dangerous content he campaigns against, including child sexual abuse and networks of so-called bot accounts deliberately designed to mislead, is not being addressed as before, they say.
Twitter has so far tried to protect itself not only from individual trolls but also from so-called “influence operations” – state-sanctioned campaigns aimed at undermining democracy and targeting dissidents and journalists.
Ray Serrato worked on a team that specialized in these operations. He left the company in November because he felt there was no clear vision to protect users under the new leadership. He says his team identifies suspicious activity like this “on a daily basis.” Now his team has been “decimated” and exists in a “minimized capacity”.
picture description,
Ray Serrato, who specializes in fighting state-sponsored disinformation, says his team has been “decimated”.
“Twitter might have been the haven where journalists would speak out and criticize the government. But I’m not sure that will be the case anymore.”
“There are a number of key experts that are no longer on this team that would have covered specific regions or threat actors from Russia to China,” he tells me.
Another insider we call Rory is also very concerned about this leak of knowledge — and how it appears to be undermining a priority of Musk’s by preventing pedophiles from using Twitter to groom victims and share links to abuse . Rory was until recently employed as part of a team fighting against the sexual exploitation of children [CSE].
His team would identify accounts sharing abusive content about children, which at worst escalates to law enforcement. Before the takeover, such content was a big problem, he says – and he was already concerned about a lack of staff.
“Any day you would be able to identify this type of material,” he says.
But his team was downsized from 20 to about six or seven shortly after the takeover, he says. In his opinion, that’s not enough to keep the workload under control.
Rory says that neither Mr Musk nor any other member of the new management made contact with him and his old team, who together had years of experience in the field, prior to his departure.
“You can’t take over a business and suddenly think you have knowledge… to deal with [Child Sexual Exploitation] without having the experts on site,” he says.
Twitter says it removed 400,000 accounts in one month alone to “make Twitter more secure”. But Rory is concerned there are now fewer people with the knowledge to effectively escalate concerns about this content with law enforcement.
“You can definitely ban hundreds of thousands of accounts in a month. But when reporting this content [to law enforcement] dropped, then it doesn’t really mean anything, and most users who had their accounts banned would just set up a new account anyway.”
He adds that offending users can then easily set up new accounts, while banned profiles are welcome back on Twitter.
I wanted to ask Elon Musk about the acquisition, his vision for Twitter, and his take on the reality. I’ve tried contacting him via email, tweets, and even a Twitter “poll.” This wasn’t a real poll, but Mr Musk used those votes to make decisions about the future of Twitter, and I was hoping it might get his attention. More than 40,000 users voted and 89% said Mr. Musk should interview me. I had no answer.
picture description,
Elon Musk hasn’t officially responded to Panorama — but he tweeted after we published this article
Twitter and Musk have yet to formally respond to the BBC Panorama investigation.
I was told that Twitter’s entire communications team was either terminated or fired. Twitter’s policies, which are publicly available online, say that “defending and respecting the voice of the user” remains one of its “core values.”
However, Musk tweeted about our post after it was posted, saying, “Sorry I turned Twitter from a nurturing paradise to a place that…has trolls.”
Were you affected by the issues raised in this story? Share your experiences via email [email protected].
Please provide a contact number if you are willing to speak to a BBC journalist. You can also get in touch with us in the following ways:
If you are reading this page and cannot see the form, you must visit the mobile version of the BBC website to submit your question or comment, or you can email us at [email protected]. Please include your name, age and place of residence with each submission.