Twitter to roll out a new feature “safety mode” to curb abuse.

September 02, 2021 | 2 minutes read

Twitter is launching a new feature in an attempt to curb and contain the abuse and trolling on the platform which has increased exponentially lately.

Safety mode will pick accounts that spew hatred on the platform or those bombarding people with uninvited comments and messages and will block them for seven days.

The feature will be activated once turned on and it will tackle accounts automatically making it easy for people to sit back and relax.

It is yet to be tested on a small group of users before it will be available publicly.

The feature can be turned on in the settings, accounts will be assessed by identifying a relationship between the user and the replier and if the interaction is frequent then the account will not be automatically blocked. 

Katy Minshall, head of Twitter UK Public Policy, said: “While we have made strides in giving people greater control over their safety experience on Twitter, there is always more to be done.

“We’re introducing Safety Mode; a feature that allows you to automatically reduce disruptive interactions on Twitter, which in turn improves the health of the public conversation.”

Like other social media platforms, Twitter also relies on a mix of automated and human moderation.

The number of human moderators hasn’t been revealed officially but a 2020 report by New York business school NYU suggested that it have around 1500 human moderators for 199 million Twitter users on a daily basis. 

A recent study on hate speech produced by Facts Against Hate on behalf of the Finnish found out Twitter to be the “worst of tech giants” when it came to hate speech on the platform. 

The answer, according to study author Dr. Mari-Sanna Paukkeri, is to utilize artificial intelligence systems which have been trained by humans.

“There are so many different ways to say bad things, and it is rocket science to build tools that can spot these,” she said.

She added that it is insufficient to reply on identifying a few sentences and graphics as hate speech and it is important to widen the spectrum.

Twitter is planning to take down misinformation alongside abuse and trolls. In August, partnered with Reuters and Associated Press to debunk false news and information and to control its spread.

It has previously introduced Birdwatch, a community-moderation system, which allowed volunteers to label tweets they found to be inaccurate.

 It still has to come with features that can actively identify misinformation and control its spread timely.