YouTube tests new viewer notes feature to combat misinformation

YouTube is testing a new feature that lets viewers add “Notes” under videos to provide additional context and information, the company revealed exclusively to TechCrunch.

This feature is akin to Community Notes on X (formerly Twitter).

Google-owned YouTube explained that these notes could be used for purposes such as indicating when a song is a parody or alerting viewers if old footage is being presented as current news.

This rollout is timely, arriving during a crucial U.S. election year. With misinformation already a major concern in the 2020 presidential election, the potential for misinformation in the 2024 election is even higher due to the proliferation of generative AI.

YouTube’s new feature appears to be a proactive measure to combat misinformation on its platform.

The experimental feature will initially be available on mobile devices in the U.S. and in English.

YouTube expects some trial and error during the pilot phase, anticipating notes that may not perfectly match a video or that may contain inaccuracies.

The company is committed to learning from this phase and is seeking feedback from both viewers and creators regarding the quality of the notes.

Only a limited number of users will be invited to write notes during the test phase, and these users must have an active YouTube channel in good standing.

U.S. viewers will begin seeing notes on videos over the next few weeks and months. During the pilot, third-party evaluators will assess the helpfulness and accuracy of the notes. YouTube will use this feedback to train its systems.

If evaluators deem a note helpful, it will appear under the video. Viewers can then rate the note as “helpful,” “somewhat helpful,” or “unhelpful,” and provide reasons for their ratings, such as citing high-quality sources or a neutral tone.

YouTube will employ an algorithm to decide which notes to publish based on these ratings. The algorithm will prioritize notes that are widely regarded as helpful.

For example, if many users who previously rated notes differently now find a particular note helpful, YouTube is more likely to display that note.

The system is designed to continuously improve as more notes are submitted and evaluated across various video types. YouTube will use the insights gained during this phase to determine if a full rollout of the feature is warranted.