WASHINGTON • Twitch, the live-streaming platform, has said it is suspending President Donald Trump's channel for "hateful conduct", in what appears to be the first deliberate suspension of one of his social media accounts.
The site, which is owned by Amazon, said on Monday that two recent streams on Mr Trump's channel violated its rules.
One stream was of a rebroadcast of a 2015 campaign event in which he made comments about Mexico sending drugs, crime and rapists over the border. The other was of his recent rally in Tulsa, Oklahoma, where he talked about a "very tough hombre" breaking into a woman's house at 1am.
"Hateful conduct is not allowed on Twitch," said a spokesman. "In line with our policies, President Trump's channel has been issued a temporary suspension from Twitch for comments made on stream, and the offending content has been removed."
It is unclear how long the suspension would last.
Twitch is not one of Mr Trump's top social media channels. His channel began streaming on the service in October, amassing over 125,000 followers and 113 streams, compared with his more than 83 million followers on Twitter.
In recent months, some tech companies have become more proactive in handling speech issues by Mr Trump and his supporters.
Twitter has started adding labels to some of the President's tweets; Snap has said it will stop promoting his Snapchat account; and Reddit said on Monday it would ban "The-Donald" community, which had been a highly influential digital gathering place for Mr Trump's acolytes. But unlike those efforts, Twitch has directly clamped down on the President himself.
The only other time Mr Trump had one of his social media accounts suspended was by accident in 2017, when his Twitter account was unexpectedly disabled by a rogue contractor who was leaving Twitter that day.
One company that has maintained that it does not want to police free speech is Facebook.
Last week, it announced that it would expand its hate speech policies and label posts from political figures who violate rules as "newsworthy". But the labels, which do not explain what is inaccurate or hateful about a post, fall short of what Twitter and other companies have done.
SMALL PLAYERS IN THE LEAD
If smaller platforms start taking more aggressive or harder action on what they consider harmful content or on the disinformation side - will that end up pressuring the larger platforms to do more as well?
MS CINDY OTIS, a disinformation expert and senior fellow at the Atlantic Council's Digital Forensic Research Lab
Twitch's suspension comes as the platform, which is popular with gamers, is under fire for other instances of hateful rhetoric. Streamers have accused it of allowing racist and sexist comments to thrive unchecked, and it said last week it would permanently suspend some users after a torrent of sexual harassment and assault allegations rocked the video game industry.
Ms Cindy Otis, a disinformation expert and senior fellow at the Atlantic Council's Digital Forensic Research Lab, said Twitch's suspension might pressure other companies to ratchet up their actions.
"You have to sort of wonder, if smaller platforms start taking more aggressive or harder action on what they consider harmful content or on the disinformation side - will that end up pressuring the larger platforms to do more as well?" Ms Otis asked.
The actions are likely to revive charges by conservatives that social media platforms are suppressing and censoring their speech.
Professor Whitney Phillips, who researches disinformation at Syracuse University, said the moves are "definitely going to be weaponisable by people on the far right who can point to this" and say that online platforms are biased against conservatives.