Take the latest VB Survey to share how your company is implementing AI today.

Twitter has placed a warning label over a tweet President Trump posted in response to Minneapolis riots following the killing of George Floyd. In the tweet, which was cited as “glorifying violence,” Trump seemed to call for violent retaliation against protestors if looting continued.

The decision is Twitter’s second riposte against Trump this week, as the social media platform previously placed a fact-checking label on a tweet about mail-in ballots. Trump responded to the earlier move by signing an executive order calling for a review of legal protections for speech on platforms such as Twitter and Facebook.

Twitter’s latest measure goes one step further by hiding the original tweet under a warning label. Users can click the label to see the tweet, but they cannot like it or retweet it.

“We’ve taken action in the interest of preventing others from being inspired to commit violent acts, but have kept the tweet on Twitter because it is important that the public still be able to see the tweet, given its relevance to ongoing matters of public importance,” reads a tweet from Twitter’s public relations team.

The killing of George Floyd, an African-American man, by Minneapolis police has sparked widespread protests in the city. Protests escalated in places to include people storming stores and seizing items. Some businesses have also been burned, and protestors took control of a police precinct and set it on fire.

While city leaders have called for calm and a return to peaceful protest, Trump took a more incendiary tone by threatening to send in the National Guard. Then he called the protestors “thugs” and wrote: “when the looting starts, the shooting starts.”

Many people noted that the phrase seemed to reference a former Miami police chief who used strong-arm tactics against minority protestors in the 60s. The language proved to be enough for Twitter to take action.

In its policy against violence, Twitter says users “can’t glorify, celebrate, praise, or condone violent crimes, violent events where people were targeted because of their membership in a protected group, or the perpetrators of such acts.”

Beyond the speech itself, Twitter notes that the president’s language could potentially incite others to engage in violent action: “We have a policy against content that glorifies acts of violence in a way that may inspire others to replicate those violent acts and cause real offline harm.”

In this case, Twitter believed that Trump’s “shooting starts” phrase crossed the line. While offending content could be removed, Twitter also highlighted its public interest exceptions policy:

“At present, we limit exceptions to one critical type of public-interest content — tweets from elected and government officials — given the significant public interest in knowing and being able to discuss their actions and statements. As a result, in rare instances, we may choose to leave up a tweet from an elected or government official that would otherwise be taken down.”

Twitter appears to have taken action around 4 a.m. Eastern, so Trump has probably not yet seen the obscuring label. However, the president’s ongoing grievances with social media platforms are likely to reach new heights.

Update at 5:42 a.m. Pacific: President Trump and the White House have responded to Twitter’s move.

Trump accused Twitter of political bias:

The official White House Twitter account sought to circumvent the warning label by reposting the same tweet:

From the White House Deputy Chief for Communications:

https://twitter.com/Scavino45/status/1266343153466060803