What to know:
- Bluesky announced it will not use user content to train generative AI, contrasting with X’s new policy allowing AI training on user posts.
- The platform has seen a surge in users, reaching over 17 million users with more than 1 million new signups in the past week.
- While Bluesky uses AI for content moderation and its Discover feed, it emphasizes these are not generative AI systems trained on user content.
In a move that sets it apart from its competitors, Bluesky has declared it will not use your posts to train generative AI models. The announcement comes at a pivotal moment when X (formerly Twitter) has just updated its terms of service to allow AI training on user content.
The rapidly growing social platform made this declaration as it experienced a significant surge in its user base. Over one million new users joined Bluesky in just 24 hours, pushing its total user count past 17 million. This surge coincides with a mass exodus from X, particularly following recent political developments and policy changes.
Despite Bluesky’s commitment not to use posts for AI training internally, there’s a catch. As The Verge reports, the platform’s current technical setup doesn’t prevent major AI companies from potentially scraping user posts. As an open and public social network, Bluesky’s robots.txt file currently allows crawlers from companies like Google and OpenAI to access its data. The company says it’s actively working on solutions to ensure outside organizations respect user consent.
Bluesky also acknowledges that while it does employ AI technology, its use is limited to specific functions. The platform uses AI to assist with content moderation, helping to shield human moderators from harmful content and to power its Discover feed’s algorithms. However, the company emphasizes that none of these systems are trained using user-generated content.
For users seeking detailed information, Bluesky has made all its policy documentation easily accessible in one central location, including its terms of service and community guidelines.
This stance particularly resonates with artists and creators who have found a new home on Bluesky. Many content creators have expressed concerns about their work being used to train AI systems on other platforms, making Bluesky’s commitment particularly significant.
The timing of this announcement is noteworthy, as it coincides with X’s implementation of new terms that explicitly allow the platform to use public posts for AI training. This contrast in approaches to user data and AI development highlights the growing divide in how social media platforms handle user privacy and content rights.
Discussion