Elon Musk’s Twitter relies heavily on automation to moderate content, according to the company’s new head of trust and security, amid a reported increase in hate speech on the social media platform.
Ella Irwin told the Reuters news agency that Musk, who took over the company in October, was focusing more on using automation, arguing that Twitter had made the mistake of using time and labor-intensive human reviews of harmful content in the past.
“He has encouraged the team to take more risks, act quickly and secure the platform,” she said.
In terms of child safety, Irwin said Twitter had shifted to automatically removing tweets reported by trusted figures with a track record of accurately flagging harmful posts.
Twitter is also more aggressively restricting abusive hashtags and search results on topics such as child exploitation, regardless of the potential consequences for “benign use” of those terms, she said.
“The biggest thing that has changed is that the team is fully capable of moving quickly and being as aggressive as possible,” Irwin said.
Her comments come after investigators reported a spate of hate speech on the social media service after Musk announced an amnesty for accounts suspended under the company’s previous leadership that had not broken the law or engaged in “blatant spam.”
The company has faced sharp questions about its ability and willingness to moderate harmful and illegal content since Musk cut half of Twitter’s workforce and issued an ultimatum to work long hours, resulting in the loss of hundreds of additional employees. .
On Friday, during a meeting with French President Emmanuel Macron, Musk pledged to “significantly strengthen content moderation and protect freedom of expression.”
Irwin said Musk encouraged the team to worry less about how their actions would affect users’ growth or revenue, and said security was the company’s number one priority. “He stresses that every day, several times a day,” she said.
Researchers say the number of tweets containing hateful content on Twitter rose sharply in the week before Musk tweeted on Nov. 23 that hate speech impressions or opinions were declining. Tweets containing words anti-black that week were three times as many as in the month before Musk took over, while tweets containing a gay slur were up 31%, a study by the Center for Countering Digital Hate found.
Irwin said layoffs did not significantly affect full-time employees or contractors working in what the company called its “health divisions,” including in “critical areas” such as child safety and content moderation.
She said Twitter has removed about 44,000 accounts involved in child safety violations, working with cybersecurity group Ghost Data.
Twitter also restricts hashtags and search results often associated with abuse, such as those aimed at looking up “teen” pornography. Previous concerns about the impact of such restrictions on allowed uses of the terms have dissipated, she said.
Using “trusted reporters” was “something we’ve discussed on Twitter in the past, but there was some hesitation and frankly just some delay,” Irwin said.