San Francisco (AFP) – Less than two minutes. It took time for the Twitch platform to interrupt the live broadcast of the Buffalo shooting on Saturday, which did not prevent the extracts from being broadcast.
Despite advances in technology, preventing live broadcasts of violent images is still a challenge, especially since the legal framework is almost non-existent.
“If (platforms) offer live, they expose themselves to a certain number of rapes, murders, suicides and other crimes,” argued Mary Ann Franks, a law professor at the University of Miami. “It’s a lot of parts.”
The 100-second time Twitch took to identify and offline the “livestream” of Paton Gendran, who killed ten people at a New York supermarket on Saturday, testified to the increased responsiveness.
In March 2019, it took Facebook 17 minutes to stop Brenton Taranto attacks on two mosques in Christchurch, New Zealand, killing 51 people.
Worse, in October 2019, again in Twitch, the author of a synagogue attack in Halle (Germany) was able to simultaneously upload his journey for 35 minutes before being disconnected.
If major social networks claim to track these videos with artificial intelligence, then with the help of dedicated teams, images can be quickly downloaded, edited and posted to other sites ready to host them.
On Wednesday, a site offered several quotes from the shooting, including a 90-second long one, which has been viewed nearly 1,800 times since Sunday, which was removed later in the day.
The introduction of non-direct violent content is almost systematic, due to the lack of applicable text to prevent it.
“In the United States, posting a live video (from Buffalo) is not illegal,” said Ari Ko, of the TechFreedom think tank. “It does not fall into a form of expression that is not protected by the US Constitution.”
For more established sites, finding this violent content, often in a new format, brought back online with a new title, is a perpetual search, a Facebook spokesperson explained.
Twitter has a policy of suspending the accounts of suspected attackers and allows “removal of tweets that spread the manifesto or content” produced by these authors.
In an exchange with reporters Tuesday, Guy Rosen, vice president of Integrity of Meta Content, explained that filters need to be carefully calibrated to avoid correction of related images, such as news videos or testimonials from people condemning the attack.
Despite the investment of large platforms, it is impossible, by definition, to deter anyone from live broadcasting, even from a few seconds of violent action.
“The main problem is when technology groups make decisions for the general public,” said Mary Ann Franks, who said “live is a tool whose utility transcends difficulties.”
However, at a time when video is a major driver of the growth of social networks, “live” does not seem to be under threat, especially with the rise of direct sales, a new mode of online commerce that has wind in its sails.
Twitch, Facebook, Twitter, Instagram or YouTube all offer live, with Snapchat being the only exception.
In a statement blaming him, the accused Buffalo shooter described the possibility of being able to broadcast his attack live as a “motivation” factor.
In the absence of federal law, it is up to the American states to take the initiative.
In Texas, a law passed last September aims to ban social networks from dismissing content based on their author’s “perspective.” It has been criticized for limiting restraint and allowing violent messages or images to spread.
“The recent (buffalo) tragedy makes it clear that this is not a biased issue,” said Matt Schroers, president of the Computer and Communications Industry Association (CCIA).
“Interrupted life and death consequences when the industry wants to go after bad actors.”
© 2022 AFP