Harry and Meghan, the Duke and Duchess of Sussex, have long cited social media toxicity as a factor in their decision to step back from royal life. And now, an analysis of more than 114,000 tweets about the couple has revealed a coordinated campaign of targeted harassment of Meghan on Twitter — and the 83 accounts responsible for approximately 70% of the negative and often hateful content.
On Tuesday, Twitter analytics service Bot Sentinel released a report examining Twitter activity related to the Sussexes and found that the majority of the hate and misinformation about the couple originated from a small group of accounts whose primary, if not sole, purpose appears to be to tweet negatively about them. Bot Sentinel’s analysis also revealed a level of sophistication and coordination between the accounts, who use their combined 187,631 followers to fuel a campaign of negativity against Harry and Meghan.
A Twitter spokesperson told BuzzFeed News on Tuesday that they are “actively investigating the information and accounts referenced in this report — we will take action on accounts that violate the Twitter Rules.”
The couple currently does not have a social media presence on any platform (neither does Archewell, their charitable foundation and production company). Meghan told Fortune in October 2020 that she hadn’t been on social media in a long time “for [her] own self-preservation.” In a January interview, Harry said that he and Meghan “will revisit social media when it feels right for us — perhaps when we see more meaningful commitments to change or reform.”
Bot Sentinel is a crowdfunded website and browser plug-in created in 2018 to analyze the activity of Twitter accounts and identify those that violate the platform's terms of service. It evaluates an account and assigns it a score based on the likelihood that the account is either run by a bot or by someone engaging in targeted harassment, trolling, or other behavior banned by Twitter’s rules. Past analyses by Bot Sentinel include the right-wing campaign to discredit the results of the 2020 American election, accounts behind COVID-19 disinformation, and online abuse toward UK football players after the team lost in the Euro 2020 final.
Bot Sentinel CEO Christopher Bouzy told BuzzFeed News that this anti-Meghan Twitter campaign is unlike anything he and his team have ever seen before.
“There’s no motive,” he said, comparing the anti-Meghan campaign to other disinformation and harassment campaigns on Twitter such as the #StopTheSteal movement to overturn the results of the 2020 US presidential election or the campaign to remove actor Amber Heard from the upcoming Aquaman sequel as a result of abuse allegations made against her by ex-husband Johnny Depp. “Are these people who hate her? Is it racism? Are they trying to hurt [Harry and Meghan’s] credibility? Your guess is as good as ours.”
In the report, Bot Sentinel includes examples of tweets from this group of 83 users that violate Twitter’s terms of service against targeted harassment. BuzzFeed News received an advance copy of the Bot Sentinel report and the list of users identified within it and found hundreds of additional tweets that violate Twitter’s rules.
Twitter’s terms of service forbid abusive behavior and engagement “that harasses or intimidates, or is otherwise intended to shame or degrade others” and prohibit social coordination between accounts that results in harm to others or that encourages dogpiling.
Bouzy emphasized that the negative Twitter activity is not fueled by automated bot accounts, but real accounts run by humans.
“This campaign comes from people who know how to manipulate the algorithms, manipulate Twitter, stay under the wire to avoid detection and suspension,” he said. “This level of complexity comes from people who know how to do this stuff, who are paid to do this stuff.”
Bouzy said the users targeting Meghan and Harry are operating “in a more clever way than we normally see.” He said their hateful tweets are mixed in with tweets that do not violate Twitter’s terms of service, making these accounts harder for Bot Sentinel to automatically detect.
Bouzy told BuzzFeed News that it’s easier for these single-purpose anti-Sussex accounts to also avoid detection because they pair their negative content about Harry and Meghan that violate the terms of service with positive comments about other members of the royal family, particularly Harry’s brother and sister-in-law, the Duke and Duchess of Cambridge.
For this report, Bot Sentinel used hashtags and specific royal or Sussex-related keywords, such as #HarryandMeghan and #MeghanMarkle, to identify a sample group of 114,000 tweets.
Bot Sentinel identified 55 “primary accounts” responsible for the vast majority of original and derivative negative content against Meghan. Another 28 users were flagged as “secondary accounts” that primarily amplify the tweets of the first group. According to Bot Sentinel’s analysis, these 83 accounts are responsible for 70% of the negative hate content targeting the couple on Twitter. In the report, Bot Sentinel estimates that the accounts have a combined potential reach of at least 17 million Twitter users.
BuzzFeed News examined the timelines of these 55 primary accounts and found tweets with clear racial undertones, one calling for Meghan’s death, as well as posts claiming that she faked her pregnancy and that her children were either born via surrogate or are not real.
“There’s a difference between free speech and literally harassing someone,” Bouzy said.
The report states that 40% of the primary accounts listed have been previously suspended by Twitter, and Bouzy said that most of the primary accounts engage in behavior specifically to avoid suspension, such as making their accounts private or temporarily deactivating them, only to return days later.
“What these accounts are doing — they’re really flying under the wire,” Bouzy said. “They’re right at the edge. They’re doing things our technology isn’t going to catch.”
Once these primary accounts were identified, Bot Sentinel looked for accounts that would quickly share their content and identified 28 separate anti-Meghan-and-Harry accounts whose Twitter activity showed hundreds of retweets from accounts in the first group.
Further analysis of followers, retweets, mentions, and other interactions between these 83 identified accounts indicated a high level of coordination between them. Many of the tweets from these accounts violate Twitter’s policies against coordinated harmful activity, which, among other things, prohibit accounts from encouraging their followers to engage in targeted harassment. One of the primary accounts in the list, @duchofnarsussex, has openly recruited individuals to join an anti-Meghan-and-Harry “intelligence agency” to “end the House of Monte-Shite-SO” (the Sussexes live in Montecito, California). The person behind the account, who declined to provide their name, told BuzzFeed News that it was created “for fun” and that “these things are being taken as literal and they are nothing more than jokes.”
“These primary and secondary accounts are pushing [hate and disinformation] out, and then it’s being taken over by organic hate,” Bouzy said. “You have the ringleaders, then the secondary accounts, and then an outside network that also amplifies and pushes a certain narrative.”
Bouzy said that although he was aware of the campaign against Meghan before this report, the impetus for the investigation came after he shared a positive tweet about the Sussexes that resulted in him personally receiving hateful messages from anti-Meghan accounts.
When Bouzy tweeted that Bot Sentinel would be launching an investigation into anti-Sussex accounts on Twitter, he became the target of a full-fledged harassment campaign unlike anything he or his team had experienced before, he told BuzzFeed News.
While every investigation pursued by the company garners hate tweets and sometimes even death threats, he said, the targeted harassment from anti-Meghan accounts was “different” and “personal.”
Bouzy said that accounts spammed his timeline and mass-reported him to Twitter (encouraging their followers to do so as well), all while sharing false allegations about him and Bot Sentinel in an attempt to preemptively discredit the report’s findings.
BuzzFeed News made a request for comment to Twitter about the accounts identified in this report on Monday. By Tuesday morning, 4 of the 83 accounts had been suspended and 6 had made their accounts private, a tactic that Bouzy said is often used by abusive accounts to prevent other users from reporting their tweets.
“This is a game they play with Twitter, and they know exactly what they’re doing,” Bouzy said.