Apps & Consumer
X open-sources its ranking algorithm and lets users check for 'shadowbans'
X is open-sourcing its "For You" ranking algorithm and adding a tool that lets users check if their account or posts were affected by its ranking systems.
X said Thursday it is significantly expanding its open source codebase — including the app’s “For You” algorithm and core ranking engine — and adding a feature that lets users see if their account or posts have been affected by any of its ranking systems. The company is publishing the source code for the “For You” timeline on GitHub under the Apache v2 license, now including model configuration, filtering, and the parameters used to weight different signals — key to understanding which posts actually get shown — making the codebase roughly 10 to 15 times larger than before. “This is the kind of thing that I think people will be fairly shocked that we are releasing,” X VP of Product Keith Coleman told TechCrunch.
Alongside the repository, X is rolling out a transparency tool on an “Under the Hood” settings page that lets users who have posted 10 or more times in the past month download their aggregate stats as a JSON file, showing whether any labels were applied to their account or posts over the past month. Non-technical users can feed that file to an LLM alongside X’s GitHub repo for an interpretation. The tool will initially roll out to a pilot group of accounts at least a year old.
Ahead of launch, X previewed the codebase to external researchers, who were able to train and run X’s Phoenix scoring system using the open source code — a milestone Coleman called significant for the company’s transparency efforts. Developers will also be able to submit pull requests for X engineers to consider, though not everything will make the cut. A few systems are excluded, including those that use Grok to predict whether a post violates X’s rules, to keep bad actors from using that information to evade moderation and flood the network with spam.
The move addresses long-running concerns about how X’s algorithm shapes politics, elections, and the spread of misinformation. The platform, now owned by a trillionaire who helped elect President Trump, faced accusations even before Musk’s takeover — Republicans in Congress once alleged the platform, then Twitter, leaned left and “shadowbanned” their posts, which Twitter consistently denied. X has become less transparent in other respects since going private under Musk, no longer required to report user metrics, revenue, or government takedown requests to the SEC — though monthly active user counts are public again now that X has merged with SpaceX.
Why it matters
Making the ranking code auditable gives outside researchers and governments a way to check X’s claims about fairness directly, rather than relying on the company’s word — a bar few other major platforms have agreed to meet.