• On June 3, in response to press inquiries about its fact-checking of President Trump’s tweet, clarified that it will not tackle all misinformation on its platform but will instead focus on posts with the “highest potential for harm”. Twitter will “provide context, not fact-checking”; “prioritize based on the highest potential for harm”; and “focus on manipulated media, civic integrity, and COVID-19”. Flagged tweets will link to a page that will show “factual statements,” “counterpoint opinions” or “ongoing public conversation around the issue.” Likelihood, severity and type of potential harm — along with reach and scale — factor into their decisions.[1]

  • On May 26, for the first time applied a fact-checking notice to tweets from President Trump: two tweets about the potential for fraud involving mail-in ballots. The move was based on a policy announced May 11 (below) to apply fact-checking labels about the coronavirus and other disputed issues subject to misinformation, including the election. This marked the first time Twitter has applied the fact-checking label to a message about non-Covid news - or to one from President Trump.[2]

  • On May 20, announced it is testing a new feature that allows a user to select, tweet by tweet, who could respond to their post. A new permissions button that appears in the bottom left corner of a tweet will allow users to choose whether they want any one (the default), "people you follow", or "only people you mention" to weigh in (people can still quote-tweet but it won’t be in the original user’s feed). While created in part to deal with bullying and harassment, it may be relevant to the spread of misinformation: it may prevent people from correcting misinformation in a tweet, or turn reply threads into partisan zones, reinforcing the original misinformation.[3]

  • On May 11, announced in a company blog that the company would start affixing labels and warnings to tweets containing information about COVID-19 that goes against the advice and knowledge of public health experts. That will include tweets where the misinformation is contained in media, like videos and images. The company said, “during active conversations about disputed issues, it can be helpful to see additional context from trusted sources”.Twitter says it will use "internal systems to proactively monitor content related to COVID-19," as well as "rely on trusted partners" for assessing the risk level of tweets. These labels will link to a Twitter-curated page or external trusted source containing additional information on the claims made within the Tweet. The announcement broadens a policy to label or remove misinformation that could cause physical harm or widespread panic to content where harm is not as obviously imminent.[4] Most coverage of the announcement focused on a response to questions from reporters in which executive Yoel Roth clarified that the company's new policy would apply to the tweets and content of world leaders — including President Trump.

  • On May 8, pushed back on an assertion from the US State Department Friday that it was "highly probable" that the Chinese government coordinated networks of Twitter accounts to disseminate disinformation related to the coronavirus outbreak, saying their initial review of the accounts in question does not support the government's claims. An initial review from Twitter of more than 5,000 accounts turned over to them by the State Department found that numerous accounts belong to government entities, nongovernmental organizations, and journalists.[5]

  • On April 22, broadened its guidance on “unverified claims that incite people to engage in harmful activity, could lead to the destruction or damage of critical 5G infrastructure, or could lead to widespread panic, social unrest, or large-scale disorder”. Material of this nature may be considered in violation of Twitter’s policies.[6]

  • On April 8, briefly locked the account of online personalities Diamond and Silk over a tweet that violated the company’s rules against coronavirus misinformation.[7]

  • On April 6, Ad Age reported (via Ad Exchange) that Twitter had lifted its ban on coronavirus-related advertising. Twitter will allow marketers to feature their pandemic responses in paid tweets. Twitter initially banned COVID terms to forestall misinformation, but now feels that “the messaging that brands and businesses can provide to the world … are going to be positively received”.[8]

  • On March 23, Daily Beast and The Hill reported that Twitter was leaving up misinformation about U.S. origins of the virus attributed to the Chinese government.[9] Twitter confirmed only that “official government accounts engaging in conversation about the origins of the virus and global public conversation about potential emergent treatments will be permitted, unless the content contains clear incitement to take a harmful physical action”. They also acknowledged “we will not be able to take enforcement action on every Tweet that contains incomplete or disputed information about COVID-19”.[10]

  • On March 20, announced it will accelerate its notoriously opaque process for verification (the blue checkmark) for accounts that are providing credible updates around COVID-19. They are prioritizing verification for Twitter accounts that have an email address associated with an authoritative organization or institution. They explained steps academics and others who work for public health organizations or academic institutions can take in order to ensure their accounts can be verified.[11]

  • On March 19, banned paid ads and commerce listings for masks, hand sanitizer, and surface disinfecting wipes to help protect against inflated prices and predatory behavior. Will be ramping up automated enforcement for ads and commerce the week of March 23. If Twitter sees abuse around these products in organic posts, they will also be removed.[12]

  • On March 19, announced it will implement a new policy on "synthetic and manipulated media," primarily focused on the 2020 election. When users scroll through posts, they may see a new labeling system: a blue exclamation point and the words "manipulated media" underneath content the platform believes to have been tampered with or deceptively shared (deepfakes,"cheepfakes", low-tech editing).[13]

  • On March 18, expanded its safety rules; it will no longer allow content that could place people at a higher risk of transmitting COVID-19, including "Denial of expert guidance," "Encouragement to use fake or ineffective treatments, preventions, and diagnostic techniques," and "Misleading content purporting to be from experts or authorities."[14] It will also take action against posts alleging that any particular group or nationality is more or less susceptible to coronavirus.[15]

  • Expanding its search feature to provide information about COVID-19 from credible sources. In January, it launched the COVID-19 Search Prompt in 64 countries and 20 languages in partnership with the CDC.[16]

  • Increasing its use of machine learning and automation to act on potentially abusive and manipulative content, (i.e., using tech to make more enforcement calls). Due to the risk of errors, will not permanently suspend any accounts based solely on automated enforcement systems; will layer in human checks.

  • Instituting a global content severity triage system to prioritize rule violations with biggest risk of harm and reduce the burden on people to report them.

  • Executing daily quality assurance checks on content enforcement processes to ensure agility in responding to this rapidly evolving, global issue.

  • Engaging partners around the world to ensure escalation paths remain open and urgent cases can be addressed.

  • Reviewing the Twitter Rules in the context of COVID-19 and considering ways in which they may need to evolve to account for new account behavior.

  • Dedicated COVID-19 page at top of timeline with accurate, current information.

  • Using @TwitterSafety for updates on policies[17]


Public Knowledge

1818 N Street NW
Suite 410
Washington, DC 20036