The maker of the Snapchat app is eliminating a feature known as the "speed filter" that lets users capture how fast they are moving and share it with friends, a company spokesperson told NPR.
The move is a dramatic reversal for Snap, Inc., which first introduced the feature in 2013.
Since then, Snap has defended the feature in the face of warnings from safety advocates who've argued that it encourages reckless driving. The company has also faced lawsuits from the families of those who have been injured or killed in car crashes where drivers were moving at excessive speeds, allegedly to score bragging rights on the app.
Critics of the speed filter welcomed the news, while also questioning the timing.
"Lives will be saved. Crashes will be prevented, but the lawyer in me says, 'My God, why did it take so long?'" said Joel Feldman, the co-founder of the nonprofit End Distracted Driving, one of the groups that urged Snapchat to remove the speed filter.
What exactly led Snap to scrap the feature now is unclear. Over several weeks, NPR asked Snap a series of questions about why it had stood by the speed filter for so long. A company spokeswoman told NPR, "Nothing is more important than the safety of our Snapchat community."
A month later, the same spokeswoman confirmed the speed filter would soon be gone.
The feature "is barely used by Snapchatters," she said on Thursday. "And in light of that, we are removing it altogether."
She said the company started removing the feature this week, but it may be a couple weeks before it disappears from the app for all of its 500 million monthly active users.
The feature has been linked to a number of deadly or near-fatal car crashes, often with teenagers behind the wheel.
A 2016 collision involving the speed filter left a driver in Georgia with permanent brain damage. That same year, the feature was tied to the death of three young women in a Philadelphia car accident. Also that year, five people in Florida died in a high-speed collision that reportedly involved the speed filter. In 2017, three young men in Wisconsin clocked a speed of 123 miles per hour on the feature before they crashed into a tree and died.
In response, Snap made a number of changes. It moved the speed feature from a "filter" to a "sticker" in Snapchat, lowering its prominence. It also added a "Don't Snap and drive" warning that would appear every time someone used the feature. The company also quietly capped the top speed for which a post could be shared for "driving speeds" at 35 mph. When NPR inquired about this in May, the Snap spokeswoman confirmed that the limitation had been imposed. Yet the company refused to simply drop the feature.
And the legal battles continue. Naveen Ramachandrappa, a California lawyer who sued Snap over the speed filter, wrote in a lawsuit that some teenage users of Snap believed they would be rewarded with digital prizes and trophies for recording a speed in excess of 100 miles per hour.
"Or at the very least, they want to find out if they will be so rewarded and so they drive at excessive speeds to see what will happen," he wrote.
A federal appeals court in May ruledthat the family of the young men who died in the Wisconsin crash should be able to sue Snap for being negligent in designing a product that led to foreseeable harm. Snap this week asked the trial court to toss the case out, arguing the speed filter did not cause the car accident.
A Snap spokesperson says of the some 5 billion "Snaps" users make every day, the speed feature is used in just a few thousand of them, suggesting that it was unpopular.
Irina Raicu, the director of the Internet Ethics Program at Santa Clara University, said increasingly tech companies are doing risk assessments of new products and features to try to get ahead of possible abuses.
"If you have a new tool or feature: What does it allow? What does it invite? And what does it incentivize? There are degrees of responsibility based on those three things," she said. "This Snapchat filter seems like maybe it was missing some of those conversations initially."
"Sometimes." Raicu added, "one of the most thoughtful ways to deploy a product is to never deploy it at all."