TLDR: X, formerly Twitter, has launched a pilot program allowing AI chatbots, including its own Grok and third-party systems, to draft Community Notes. This initiative aims to significantly accelerate fact-checking and combat misinformation at scale, while maintaining crucial human oversight. All AI-generated notes will undergo the same rigorous multi-perspective human review process as human-written notes before publication, ensuring accuracy and transparency.
X, the social media platform previously known as Twitter, has embarked on a significant pilot program to integrate artificial intelligence into its Community Notes feature, a user-driven fact-checking system. Launched around July 1, 2025, this initiative allows AI chatbots to draft contextual notes, with the primary goal of enhancing the speed and scalability of misinformation combat efforts.
Under the new system, AI tools such as X’s proprietary chatbot, Grok, alongside various third-party AI systems connected via an AI Note Writer API, will be capable of generating draft Community Notes. These AI-produced notes, however, will not bypass the existing human-centric review process. They must still be evaluated and rated as ‘helpful’ by a diverse group of human contributors, mirroring the stringent vetting applied to notes drafted by human users. This critical human oversight is designed to ensure accuracy, fairness, and transparency, addressing concerns about potential AI biases or the generation of misleading information.
Keith Coleman, X’s Vice President of Product and head of Community Notes, emphasized the continued importance of human judgment, stating, ‘Ultimately, the decision on what’s helpful enough to show still comes down to humans.’ He also highlighted the practical benefits, noting that ‘Humans don’t want to check every single post on X — they tend to check the high-visibility stuff. But machines could potentially write notes on far more content.’ This suggests a strategic move to leverage AI for volume while relying on human discernment for quality control.
The program is designed to be incremental, with AI bots initially focusing on posts where users have specifically requested context. Furthermore, these ‘AI Note Writers’ must ‘earn’ their ability to contribute, gaining or losing privileges based on the helpfulness ratings their notes receive from human reviewers. For full transparency, all AI-generated notes will be clearly labeled as such.
The integration of AI is expected to significantly boost the platform’s capacity to address misinformation, especially during major news events when false narratives can spread rapidly. While Community Notes have already proven effective, reportedly reducing the spread of misinformation by as much as 60%, speed has remained a challenge. This hybrid approach, combining AI for scale and humans for oversight, aims to strike a new balance, allowing for hundreds, if not thousands, more notes to be produced daily without increasing the workload on human contributors.
Also Read:
- AI Chatbots Drive New Wave of Startups Capitalizing on Evolving Search Landscape
- xAI Under Fire: Elon Musk’s Company Faces Backlash Over Unpermitted Methane Turbines in Memphis
Despite the promising potential, the initiative is not without its challenges. Experts and users alike have raised questions regarding the reliability of AI-generated content, the potential for notes to sound ‘condescending or corporate,’ and the risk of overwhelming human reviewers with a deluge of AI submissions. X is currently operating the feature in a ‘test mode’ to monitor quality and usefulness, with the first cohort of AI bots expected to begin publicly posting notes later in July, following successful pilot phases.
