Feeling Suicidal? Facebook’s AI Already Knows

Chennai, November 30: Facebook first introduced its Artificial Intelligence -based suicide prevention tools earlier this year. But until now, those tools required a user, or one of their friends, to seek help. Now the social network says the tech has advanced to the point that it can proactively intervene when it detects that someone may be at risk of self harm or suicide — even if no one else has made a report.

For many people who’ve dedicated their lives to preventing suicide, social media posts can be a precious data-set that contains hints about what people say and do before they attempt suicide. In the past few years, researchers have built algorithms to learn which words and emoticons are associated with suicidal thoughts. They’ve used social media posts to retrospectively predict the suicide deaths of certain Facebook users.


Facebook has rolled out new artificial intelligence tools that can proactively identify heightened suicide risk and alert a team of human reviewers who are trained to reach out to a user contemplating fatal self-harm.

Now, when the system identifies a post or Facebook Live broadcast “likely to include thoughts of suicide,” those posts can be routed to Facebook’s specially-trained reviewers who, in turn, can contact first respondents.

The technology, announced Monday, represents an unparalleled opportunity to understand and predict suicide risk. Before the AI tool was even publicly announced, Facebook used it to help dispatch first respondents in 100 “wellness checks” to ensure a user’s safety. The tool’s life-saving potential is huge, but the company won’t share many details about how it works or whether it’ll broadly share its findings with academics and researchers.

“The whole point of this is that our proactive detection can kick in even before something has been reported,” says Facebook VP of Product Management Guy Rosen.

That’s a big change from Facebook’s prior tools, which still relied on a user, or one of their friends, to seek help. Eliminating this step not only means the company can identify cases that may have previously fallen through the cracks, but that they can reach people who need help much faster than before.

By Parth Sharma with input from agencies