Twitter‘s new head of trust and safety said the site has become increasingly inclined to use automation to moderate content, doing away with some manual reviews and preferring to restrict the spread of certain content rather than remove it entirely.
Twitter’s vice president of trust and safety product management, Ella Irwin, said the platform would also restrict hashtags and search results that are misused in specific areas, such as child exploitation, without considering the impact of the decision on “benign uses” of those terms.
The most important thing that has changed at the company, Irwin said, is that the team is now fully empowered to move quickly and be as offensive as possible.
Irwin’s comments come amid mounting warnings from researchers about an increase in hate speech on Twitter, especially after Musk announced last November that he would issue a general amnesty for all suspended accounts, provided they were not involved in “horrible content.”
Since Musk dismissed about half of Twitter’s workforce, then gave the rest of the staff a choice between working hard for long hours or leaving, which prompted hundreds of employees to resign, doubts have risen about the company’s willingness and ability to modify content.
These suspicions led to growing fears of advertisers, who represent the main source of Twitter revenue, of harming the reputation of brands, which prompted them to temporarily stop advertising on the platform.
Yesterday, Friday, in a meeting with French President Emmanuel Macron, Musk pledged to “significantly strengthen content modification and protect freedom of expression.”
Irwin said Musk encouraged the team not to worry about how their actions would affect the number of users or revenue, emphasizing that security is the company’s top priority.
According to former employees familiar with the matter, the safety approach Erwin described reflects at least in part an acceleration of changes already planned since last year regarding Twitter’s handling of offensive content. Especially since Musk has confirmed more than once that Twitter will allow freedom of expression and will not allow freedom of access, and this requires not deleting some tweets that violate the company’s policies regarding abusive content but are prevented from appearing in places, such as the page, the main page, and the search results.
Twitter has long deployed “visibility filtering” tools for disinformation and had already incorporated them into its official policy for dealing with abusive behavior before Musk’s acquisition. This approach allows for greater freedom of expression while minimizing the potential harms associated with widespread offensive content.
Although Musk claimed last week in a tweet that “impressions of hate speech” on Twitter had decreased by a third since he took over the company, research by an organization, the Center for Countering Digital Hate, revealed that the amount of hate speech on the platform increased during the same period.