Twitter could have caused a coronavirus panic in New York. It’s not doing a lot about it

Tweets falsely claiming that New York City was going into lockdown as a result of the coronavirus swirled around Twitter Thursday evening. A “citywide quarantine” was about to come into place, one user falsely claimed.

The claims, which also circulated through text message, were so widespread that the New York Police Department and New York Mayor Bill de Blasio released statements assuring New Yorkers they were false.

“NO, there is NO TRUTH to rumors about Manhattan being quarantined. Whoever is spreading this misinformation, PLEASE STOP NOW!” de Blasio tweeted.

As of Saturday, when this article was published, a number of tweets that had falsely made the quarantine claim were still live on the social network. Asked about them by CNN Business on Friday, the company had no comment.

Responding to CNN Business’ reporting, spokespeople for de Blasio and Democratic presidential hopeful vice-president Joe Biden criticized Twitter’s response to coronavirus misinformation.

“We’re in the midst of a public health crisis. It’s wildly irresponsible for Twitter to allow misinformation to be broadcast so publicly on their platform,” Avery Cohen, a spokeswoman for de Blasio told CNN Business Saturday night.

“People’s lives are at risk. We’ve asked New Yorkers all around our city to rise to the challenge and look out for themselves and others. Twitter ought to do the same,” she added.

Michael Gwin, a Biden campaign spokesperson said, It’s absolutely shameful that in the midst of a global pandemic, Twitter is allowing accounts to post objectively false information about COVID-19 that could generate panic and risk the safety of the public.”

Twitter has done some good work since January to elevate sources of authoritative information. It says it’s cracking down on bots and has put in place additional rules for ads about the coronavirus.

But there is a gaping hole in its defenses: While, for instance, bots pushing misinformation in a coordinated way (think of one person controlling 100 Twitter bot accounts pushing false information) are against Twitter’s rules, objectively false information shared by regular users – say, individual anonymous accounts claiming that the biggest city in America is about to go into lockdown, are not.

Twitter has long-resisted fact-checking its platform.

It has done so despite knowing from past experience that the platform can be used in a situation in which there is widespread fear to spread false information that could cause people to panic.

During Hurricane Sandy in 2012 the Twitter user @comfortablysmug posted false information that news outlets, Con Edison, and the MTA, the authority responsible for the New York subway, had to correct. “BREAKING: Con Edison has begun shutting down ALL power in Manhattan,” @ComfortablySmug falsely claimed in one tweet.

The false tweet from 2012 is on the platform to this day.

“Twitter’s self-correction mechanism — rebukes and rebuttals from knowledgeable sources — shut down each rumor, but not until at least one, the flood claim, had bled widely into the television media,” Andrew Kaczynski, a CNN reporter who was then at BuzzFeed News, wrote at the time. (The “flood claim” Kaczynski was referring to was a false claim about flooding at the New York Stock Exchange.)

The person behind @comfortablysmug later apologized.

That “self-correction mechanism” is one that Twitter has relied on for a long time. Indeed, a Twitter spokesperson Friday pointed CNN Business to de Blasio’s tweet, classifying it as a corrective action, and it is true that the mayor’s tweet had a lot more retweets than the ones with the false rumors of quarantine. The spokesperson also said Twitter will take action on tweets that could directly cause physical harm.

Twitter and other companies in Silicon Valley have long resisted being the arbiters of truth. Their argument normally includes principles of freedom of speech and is rooted in the tradition of the internet being an open and unrestricted form of communication —and that’s not a weak argument. It’s also true that not fact-checking is a lot less expensive for the companies making those principled arguments.

And while Twitter isn’t fact-checking, the misinformation about coronavirus isn’t restricted to New York. On Tuesday, when primaries were happening in states including Michigan, Washington, Idaho, North Dakota, and Missouri, a fake Twitter account claiming to be “CNN Breaking News” and using CNN’s branding falsely claimed that there were thousands of confirmed coronavirus cases in those states.

The account falsely stated “CDC advises those in these states to stay home today if older than 55.”

The tweet did not go viral and Twitter did not catch it. The tweet was still live on Friday but the account had removed the CNN branding. Its name had been changed to “operation murder Joe Biden.”

Twitter suspended the account after CNN Business asked about it on Friday.

CNN’s Sarah Mucha contributed reporting