“We are testing a feature to allow you to report tweets that appear to be misleading, even when you see them”, Twitter said Tuesday, August 17, via its dedicated security account.
A handful of users in the US, South Korea and Australia can now select the option “It’s misleading” after clicking on “Report tweet”. They will then have the choice between several categories: ” health “, ” Politics “ and ” other “.
→ STORY. Social networks, the risks of parallel justice
“We are evaluating if this is an effective approach so we start on a small scale, said the San Francisco-based company. We will not and cannot respond to every report during this experience, but your contributions will help us identify trends in order to improve the speed and reach of our disinformation work ”.
A fight against disinformation deemed insufficient
Twitter, like Facebook and YouTube, is regularly criticized for not tackling disinformation enough. But the platform does not have the same resources as its neighbors in Silicon Valley, and therefore explores techniques that are less expensive than recruiting armies of moderators.
Over the years, and especially during the American election campaign and the pandemic, the social network has tightened its rules. As of March, for example, users can be banned after five reminders about vaccine misinformation. A system inspired by the one put in place regarding the elections, which notably led Twitter to ban former US President Donald Trump for repeated offenses such as incitement to violence and his messages discrediting the presidential ballot.
Moderators are responsible for determining which content violates the regulations, but the platform then indicated that it hoped to develop a system that is both human and automated to detect problematic messages. The issue of disinformation around Covid-19 and vaccines has grown to such proportions that in July US President Joe Biden even estimated that Facebook and other platforms “Killed” people by letting false information about the Covid vaccination circulate. He reconsidered his remarks to specify that false information disseminated by users could “To hurt those who listen to them” and “Kill people”.