Twitter Flags COVID-19 and Election Misinformation

Twitter Launches Misinformation Reporting Test

Twitter has initiated a new testing phase for a feature enabling users to directly report potentially misleading content encountered on the platform. This allows individuals to flag information as “misleading” to the company.

The rollout of this test began Tuesday and is currently being extended to the majority of users located in the United States, Australia, and South Korea.

How the Reporting Feature Works

Users participating in the test will find an expanded contextual menu – accessed via the three dots in the upper right corner of a tweet – which now includes a “report tweet” option. Selecting this option presents users with the ability to categorize the misleading information.

The categorization choices include “politics,” “health,” or “something else.” Further specification is available; for example, under “politics,” users can indicate if the tweet concerns elections. Similarly, under “health,” users can specifically flag misinformation related to COVID-19.

Previous Efforts to Combat Misinformation

Twitter has previously implemented methods for reporting election-related misinformation. However, these were typically temporary measures tied to specific global election events.

In 2019, the platform introduced the option to report misleading tweets concerning voting procedures, aiming to protect elections in both Europe and India.

The Goal of the New System

This new system intends to provide a means for users to highlight tweets that contravene Twitter’s established policies regarding misinformation related to elections and the ongoing pandemic.

The user reporting system will operate alongside Twitter’s existing proactive systems, which utilize both human review and automated tools to identify potentially harmful misinformation.

Currently, users will not receive notifications regarding the outcome of their reports. However, the company has indicated that such updates may be incorporated in the future.

An Experiment in Moderation

While widely available, Twitter characterizes this feature as an “experiment,” not a finalized product. The company will analyze user behavior with the new reporting tool to assess its effectiveness in identifying harmful misleading tweets.

There is no predetermined timeline for full implementation or removal of the test feature.

Concerns About Potential Abuse

Twitter does not currently express significant concern about potential misuse of the reporting feature, as it integrates directly into the existing moderation infrastructure.

Nevertheless, the ability for users to flag “misleading” tweets is anticipated to generate accusations of censorship from segments of the platform already known for disseminating misinformation.

Integration with Existing Enforcement

The reported tweets will be processed through Twitter’s current enforcement procedures, where established rules concerning health and political misinformation are applied through a combination of human and algorithmic moderation.

Prioritization of Reviews

Reported tweets will be prioritized for review based on several factors. Tweets originating from accounts with substantial follower counts, or those receiving unusually high engagement, will be reviewed promptly.

Additionally, tweets pertaining to elections and COVID-19 will receive priority, reflecting Twitter’s focused efforts in these areas.

Exploring Community-Based Moderation

This test represents Twitter’s latest attempt to leverage its user community in identifying misinformation. A more ambitious project in this vein is Birdwatch, a crowdsourced system allowing users to add contextual notes and fact-checks to tweets, similar to Reddit’s upvote/downvote system.

Birdwatch remains a pilot program, but it demonstrates the company’s interest in decentralized moderation – a considerably more complex undertaking than simply adding a new reporting option.