Twitter Will Label Posts With Misleading Information About COVID-19 Vaccines

Twitter announced Monday that it will begin applying labels to tweets that include misleading information about COVID-19 vaccines and is introducing a strike policy to curb users from repeatedly violating those rules.

In an effort to root out misinformation surrounding the coronavirus from its platform, Twitter said the labels will first be applied by human moderators. But the goal is to eventually use both humans and artificial intelligence to address content that spreads COVID-19 vaccine misinformation.

"As health authorities deepen their understanding of COVID-19 and vaccination programs around the world, we will continue to amplify the most current, up-to-date, and authoritative information," Twitter said in a blog post.

The company will start with English-language content and plans to expand to other languages over time.

In December, Twitter announced that it may require users to remove tweets that advance harmful, false, or misleading narratives about vaccines, including suggestions that immunizations and vaccines are used to intentionally control citizens.

On Monday, Twitter said that since introducing its COVID-19 guidance in March 2020, it has removed 8,400 tweets for violating policy. In that time period, Twitter said it has also "challenged" - or took steps to verify the information and authenticity - of 11.5 million accounts.

The initial guidance from March required users to remove "content that increases the chance that someone contracts or transmits the virus." That included tweets denying health expert guidance on safety, encouraging the use of ineffective treatments, and other misleading content claiming to be from authoritative sources. 

Twitter said in addition to labels, a new 5-strike policy "will help to educate the public" and "further reduce the spread of potentially harmful and misleading" information, especially for repeated violations of rules.

A tweet that includes misinformation but doesn't warrant full removal, will earn the users one strike. Twitter said it would require users to delete tweets with "high-severity" violations of policy that "invoke a deliberate conspiracy by malicious and/or powerful force."

Twitter says it may reduce the visibility of tweets and turn off likes, comments and retweets for content that is labeled with a warning message and determined to be harmful. Tweets with misleading information about the coronavirus will accrue one strike. A user will be issued two strikes if Twitter requires them to delete a tweet. 

There is no account level action for one strike, Twitter said. Once a user gets two strikes, their account will be locked for 12 hours. Similarly, three strikes also lead to a 12-hour account lock, the fourth strike leads to a 7-day account lock and the fifth strike means permanent account suspension.

YouTube has a similar strike where accounts that earn three strikes in a 90-day period are permanently removed.

This isn't the first time Twitter has placed labels on false and misleading Tweets. In November, ahead of the 2020 presidential election, Twitter said it would label all Tweets that falsely claim a victory for any candidate.

Social media companies have been ramping up efforts to remove deliberately false and misleading information about the coronavirus and COVID-19 vaccines from their platforms.

Last month, Facebook expanded the list of false claims about the coronavirus and warned that Pages, Groups, and accounts which repeatedly share debunked claims may be removed altogether.

First published on March 1, 2021 / 7:06 PM

© 2021 CBS Interactive Inc. All Rights Reserved.