Pune

X (Twitter) to Deploy AI Agents for Fact-Checking in Community Notes

X (Twitter) to Deploy AI Agents for Fact-Checking in Community Notes

X platform will now utilize AI agents for fact-checking, which will generate Community Notes. However, the final decision will rest with humans. This will enable faster mitigation of misinformation.

AI Agents: Social media platform X (formerly known as Twitter) is revamping its popular fact-checking tool, Community Notes. The company will soon begin using Artificial Intelligence (AI) agents to assess the authenticity of posts and generate neutral comments (notes) on them. This change could prove to be a significant technological and ethical shift, as this responsibility has so far been entirely dependent on human users.

What is X's new plan?

X is now asking developers to create custom AI agents that analyze the veracity of posts on the platform and write accurate, unbiased, and concise notes on them. These AI agents will have to go through the process of writing 'practice notes,' which will be reviewed by the company's team. If their quality is found to be in line with the standards, they will be allowed to write notes on public posts.

The final decision by humans

Although AI agents will prepare the notes, the final decision will remain in the hands of humans. Any note will not be displayed publicly until several users from different viewpoints consider it "helpful." This process maintains the original crowdsourcing philosophy of Community Notes, which prioritizes diversity and impartiality.

X's Community Notes lead, Keith Coleman, said in an interview that this combination of AI and human participation could prove to be "incredibly powerful." This will not only help curb misinformation but also rapidly increase the number of Community Notes, which is currently limited to hundreds per day.

Why the need for AI?

Fact-checking is a time-consuming and laborious process. It is essential to respond to rapidly spreading misinformation on social media with the same speed. According to X, this task can be done much faster, more accurately, and on a larger scale with the help of AI.

Another reason for this change is the expanding user base and the influx of fake news. Millions of posts are uploaded on the platform every day, many of which contain misleading, half-truths, or false information. Monitoring all of these is beyond the capacity of human resources alone.

How will AI agents work?

  • AI models will be trained on a large amount of text data.
  • They will be taught to understand posts, find context, and draw conclusions based on reliable sources.
  • AI will also be taught to stay away from emotional language and create neutral comments.

The company is also taking steps to ensure that AI-generated notes clearly state that they were prepared by AI, thus maintaining transparency.

The story behind: From Birdwatch to Community Notes

In 2021, Twitter launched a crowdsourced fact-checking platform called Birdwatch, in which some selected users were given the authority to add notes to public posts. These notes were visible to all users when a sufficient number of people considered them 'helpful'.

In 2022, after Elon Musk's acquisition of Twitter, Birdwatch was rebranded as 'Community Notes'. Since then, this feature has been solely operated by human users. But now, its direction is about to change with AI.

Potential benefits of this step

  • Significant improvement in the speed of fact-checking.
  • Scalability — Monitoring of a large number of posts is possible.
  • 24/7 availability — AI can work continuously without getting tired.
  • Algorithmic impartiality (if well-trained).

Potential challenges

  • Will AI be able to remain unbiased?
  • If AI adds the wrong context, who will be responsible?
  • Could this feature also be misused?

Leave a comment