The AI Note Writer API enables developers to build agents to propose Community Notes on potentially misleading content across the X platform.
Community Notes, X’s crowdsourced fact-checking system, allows contributors to add context to posts that might contain misleading information. AI agents can now join the ranks of note writers, though with due human oversight built into the process.
“Not only does this have the potential to accelerate the speed and scale of Community Notes, rating feedback from the community can help develop AI agents that deliver increasingly accurate, less biased, and broadly helpful information — a powerful feedback loop,” X explained.
The programme begins with a pilot phase and will gradually expand. Developers can immediately sign up and begin building AI Note Writers using the API, though only a select cohort will be admitted later this month, at which point AI-written notes could start appearing on X.
X emphasises that while AI will propose notes, humans retain decision-making authority. AI-generated notes will only appear on X if human Community Notes contributors from different perspectives rate them as helpful, maintaining the human-centred approach that has characterised Community Notes thus far.
The company outlined four guiding principles for the initiative. The first is openness, allowing anyone worldwide to participate using their preferred AI technology. Second is fairness, with AI-written notes held to identical standards as those written by humans. The third principle focuses on quality, requiring AI Note Writers to earn writing privileges through contributions. Finally, transparency dictates that all AI notes will be clearly identified as such to users.
Initially, the AI systems will be limited to writing notes on posts where users have specifically requested a Community Note, though X indicates this scope will likely increase over time.
To participate, developers need an X account that meets several criteria, including not being an existing Community Notes contributor, having a verified phone number from a trusted carrier, and possessing a verified email address. They must also sign up for both the X API and the AI Note Writer API.
Similar to human contributors, AI systems must earn the ability to have their notes seen by the wider community. This evaluation process involves an open-source, automated note evaluator that considers factors such as whether the note is on-topic and free from harassment or abuse.
Upon meeting admission criteria, AI Note Writers will be “automatically and randomly selected for admission,” with initial numbers kept small as X refines the system. Once admitted, AI contributors can write notes that will be shown to other contributors, receive ratings, and potentially appear across the platform if deemed helpful.
While human oversight is commendable, there are legitimate questions about whether these AI systems might amplify existing biases in fact-checking. If successful, the AI Note Writer API could create a moderation force on X that could theoretically work round the clock, but the risk is creating a false sense of comprehensive fact-checking that might not materialise.
Privacy advocates have expressed concerns about the data used to train these AI systems and whether they might inadvertently perpetuate misinformation if not properly calibrated.
For developers interested in participating, X has provided a template contributor as an open-source starting point. This template uses GitHub actions and Grok, offering a basic framework that developers can modify or replace entirely.
The launch of this API represents a notable shift in how AI might participate in content moderation efforts on platforms like X. As the pilot programme rolls out in the coming weeks, its effectiveness will likely be closely monitored by other social media companies grappling with similar challenges.
(Photo by Kelly Sikkema)
See also: Gemini CLI: Google brings AI smarts to the terminal
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Explore other upcoming enterprise technology events and webinars powered by TechForge here.
