Twitter draws flak for failing to detect suicidal messages
The demand to bar hatemongers and those spreading fake news from Twitter is increasing after two popular parody accounts — @SirJadeja and @mediacrooks — were suspended.
Following this, Supreme Court lawyer Prashant Bhushan called for action against writer-academic Madhu Kishwar, whom he dubbed “queen of fake and hate news”. When she criticised Mr Bhushan, Ms Kishwar was called out by fact-checking agencies.
Twitter pointed to the potentially harmful effects of misinformation on civic and political discourse. “We believe that no single actor can or should be the arbiter of truth. It is not possible for us to distinguish whether every single tweet from every single person is truthful or not,” said a Twitter spokesperson.
“We believe that taking down content simply because it is incorrect would ultimately undermine the open democratic debate that our platform is intended to facilitate,” he said.
Some alleged that accounts that were targeting specific parties like the Congress were being suspended. The parody account of @SirJadeja was apparently suspended twice previously.
“It is now a trend on Twitter to suspend accounts and that too selectively. When I ask Twitter for proof of tweets that are detrimental, they don’t have answers and restore it. It is being done by the same people who vouch for freedom of expression”, @SirJadeja said. Self-harm on social media is becoming a dangerous trend but Twitter is unable to help with suicide prevention as it cannot detect words used in a suicidal context.
Users tried to test if Twitter would block their account by typing the words “Kill me” but no action was taken. User reported that when they changed their name to reflect that of a woman, Twitter blocked them out.
Social media can affect emotions, self-image and self-esteem and result in self-harming and unhealthy behaviour to relieve stress. People with such intentions often find it easier to discuss their thoughts on social media, rather than in the physical world.
Given this, Twitter had announced resources for such persons, titled “Glorifying self-harm and suicide” in February.
A user, Sri Sri Ashwin, tested out the terms with his regular handle but nothing happened. When he changed his name to a female sounding one, he got a notification. “Blocked out of Twitter and the only services I am allowed to access are DMs to reach out to friends (sic),” he said in a tweet. He pointed out that there is no way for a person to reach out to experts on Twitter as cut off support.
Following this tweet, a couple of verified users tested to see if Twitter acted on such messages, only to realise that their accounts were immune. Asked about this, a Twitter spokesperson said tweets were not removed in an automated manner and they relied on bystander reporting and human review
“Context is very important when it comes to tweets using language which could pertain to self harm. For example, a user could post “Craving Mexican food and my favorite restaurant is closed...I want to kill myself ” without any intention of self harm. If we receive bystander reports for Tweets that appear to be encouraging self-harm and that don't include any other context, we will err on the side of caution and remove these tweets,” the spokesperson said.