Summary:
A few days after the Russian invasion of Ukraine, Twitter announced that it would start labeling tweets containing links from Russian state-linked media outlets, noting their affiliation. Additionally, Twitter started to reduce the circulation of this content on the platform. On that same day, a few journalists who worked for Russian media outlets, such as Russia Today and Sputnik, found that their personal accounts had been labeled as “State-affiliated media outlet, Russia.” Although the labels were removed a couple of days later, journalists impacted by this action, a few press and freedom of expression organizations, and academics criticized Twitter’s labeling move and defined it as “antidemocratic and violent,” as well as a way of stigmatizing journalists.
Background
On February 28, 2022, in the midst of the Russian invasion of Ukraine, Twitter decided to take further action in order to prevent disinformation tied to the Russian government from spreading. The company announced that it would start labeling tweets containing links from Russian state-affiliated media outlets, noting their affiliation, and would reduce the circulation of this content on the platform.
According to Yoel Roth, Twitter Head of Safety & Integrity at the time, “As people look for credible information on Twitter regarding the Russian invasion of Ukraine, we understand and take our role seriously. Our product should make it easy to understand who’s behind the content you see, and what their motivations and intentions are.” The announcement also mentioned that Twitter would apply these labels to other state-affiliated media outlets over the following weeks.
A few days before, Twitter also shared that they had been “actively monitoring vulnerable high-profile accounts, including journalists, activists, and government officials and agencies to mitigate any attempts at a targeted takeover or manipulation.”
In 2020, Twitter had already started to label accounts belonging to state-affiliated media entities, their editors-in-chief, and/or their senior staff. Twitter defined state-affiliated media as “outlets where the state exercises control over editorial content through financial resources, direct or indirect political pressures, and/or control over production and distribution.”
Between 2020 and 2022, Twitter did not recommend or amplify state-affiliated media accounts from certain countries and added labels, including to Russian-linked entities. However, as explained in 2022 by Sinéad McSweeney, VP of Global Public Policy at the time, “much of the content from state-affiliated media came from these link shares, and not just from the accounts we’d been labeling as state-affiliated media.” In his announcement on February 28, 2022, Mr. Roth underscored that they had been labeling the accounts of hundreds of global state media outlets for years, but “tweets sharing their content lacked visible context.”
The Case
On February 28, 2022—the same day Twitter announced the changes—a few journalists from Argentina, Uruguay, and Spain who worked for Russian media outlets such as Russia Today (RT) and Sputnik found that their personal accounts had been labeled as “State-affiliated media outlet, Russia.” Importantly, these journalists worked sporadically for RT and Sputnik, while also having other jobs for different media outlets in Latin America and Europe.
The journalists claimed that their accounts were not related to a media outlet and that they did not necessarily agree with their employer’s editorial line or policies. One of them qualified the labeling as “antidemocratic and violent,” and explained that he neither worked on nor mentioned anything relating to the Russian invasion on his Twitter personal account.
Representatives from various Latin American press and freedom of expression organizations expressed their concerns about Twitter’s action on the journalists’ accounts and stated that it was an unjustified penalty, “since journalists need to continue to do their jobs, and even more during times of conflict.” The organizations emphasized the importance of the right to freedom of expression and stressed that the labeling stigmatized press workers, which could lead to violence and assaults against them. Moreover, a few academics and members of civil society organizations stated that the measure was problematic, abusive, disproportionate, and unfair, and that case-specific measures such as this one may not be the correct approach. They also stated that Twitter should do the same with state-affiliated media outlets from other countries, not only Russian ones.
The Response
According to the journalists involved, the labels on their personal accounts disappeared two days after the implementation of the measure without any explanation.
Twitter’s Head of Government Affairs for Latin America tweeted about the action taken by Twitter. He explained that labels are applied to media outlets, but also to some people who are linked to them, such as editors or high-profile journalists. He also answered a few questions posed by academics and activists on the same thread.
According to one of the journalists impacted by this action, the prompt notifying him about the label provided him with an email address to learn more about the policy that had been implemented; however, it did not offer a way to appeal the decision.
Insights
This case seems to be an unintended technical error that resulted from the various actions taken to tackle the circulation of disinformation: It was related to a policy announced two years before (2020) and it was rolled back without explanation a couple of days after the action was taken. However, it is a case worth analyzing since it illustrates a relevant content moderation challenge. It involves the labeling of journalists, whose work may be fundamental for users to get trustworthy information, particularly during crises like the Russian invasion of Ukraine. Because of their role, information journalists share is strictly related to the right to freedom of expression and to the right to access to information. Thus, although Twitter policies have been implemented with the aim of providing more context and transparency during a situation of conflict, these collateral damages could be detrimental not only to journalists themselves but also to their work and, ultimately, to open and democratic debate.
Company considerations:
- What preventive measures should a company take when implementing labeling?
- What might have been a safer and scalable alternative measure?
- How can companies effectively address the disinformation problem during a specific situation of conflict without affecting policy consistency?
- How can companies offer an effective way for users to report these specific cases?
- In which cases should a company be responsible for transparently identifying who’s behind the content you see, and what their motivations and intentions are?
Issue considerations:
- What alternatives to labeling do platforms have to prevent the circulation of disinformation during crises?
- What kind of and how much context do users need when reading a post/tweet? How does this change during times of crisis?
- What could have been done to make the policies in this case clearer for users to understand their implementation and objectives? Were the policies in this case sufficiently clear for users to understand their implementation and objectives?
- Considering the potential negative impact on the right to freedom of expression and access to information, what kind of objective evidence does a platform need to show to decide to label a particular media outlet? Would that make the decision more trustworthy for users?
- Given that incorrect labeling of personal accounts may have a significant impact on a person’s social and professional life, in which specific cases would labeling be a suitable measure?
- What are the challenges of investigating content moderation issues when posts are not in English and incidents are not covered by English-language press outlets?
Written by Maia Levy Daniel, May 2023.