Twitter 'shuts down millions of fake accounts'

Share:

The suspensions and closures are part of a concerted Twitter action to clean up the platform, the newspaper said.

It is believed that many accounts are used by trolls or remote robots that abuse the service.

Twitter declined to comment on the publication's story, but said it was trying to "improve" public conversations on the social network.



Cat and mouse


Juan Guzmán, a UCL researcher who exposed hundreds of thousands of bots on social networks, said that Twitter has failed to deal with automatic tweet generators for years.

"Until recently, Twitter did not believe bots were a problem on its platform and they did not make bots," he told the BBC.

"Only after the Brexit and the 2016 elections were these robots a responsibility and Twitter and Facebook started taking them seriously".

Del Harvey, the trust and security agent of Twitter, told the newspaper that he now wrong "to maintain security" rather than to support freedom of expression at any cost.

"One of the biggest changes is the way we want to balance freedom of expression with the potential for freedom of speech to relax someone else's speech," Harvey said. 

A Twitter spokesman said the Washington Post report was not "new," adding that account lockouts are part of its "ongoing work to improve public health on Twitter."

The speaker pointed to a blog by Yoel Roth and Harvey that talked about how Twitter handled spam and bots in its service.

It has also taken action against 142,000 applications that have injected more than 130 million "low quality and spam" tweets on Twitter, he added.

Last month, Twitter also acquired Smyte, a technology company that helps people fight spam, abuse and fraud.

But Guzman, who helped develop the Astroscreen robot identification software, said Twitter was finding it increasingly difficult to detect and eliminate junk accounts.

"There is a game of 'cat and mouse' between the creators of bots and social networks, with robot manufacturers quickly updating their methods to avoid discovery," he said.

"With discoveries like Conversational Artificial Intelligence, you'll see that robots will be harder to spot in the future."

"Freedom of speech does not mean much if people do not feel safe," he said.