X opens sources its ranking algorithm, letting users check shadowban status
Social media platform X has released its post-ranking algorithm as open source, enabling users to inspect code and use AI to interpret their visibility status.
Stock photo for illustration only, not from the actual event
- X open sources its post-ranking algorithm for transparency.
- Non-technical users can use LLMs to analyze code from GitHub.
- A pilot phase is rolling out first for accounts at least a year old.
- Grok rule-prediction systems remain excluded to prevent spam.
Social media platform X has surprised the tech industry by open-sourcing its post-ranking algorithm, a major move designed to let users check if they have been shadowbanned. Company representatives remarked that this release is the kind of development people would be fairly shocked to see them put out into the open.
Even non-technical users can take advantage of this new information by dropping it into an LLM of their choice, pointing the artificial intelligence at X's GitHub repository, and asking for a direct interpretation. This bridges the gap for everyday users who want to understand complex ranking logic without reading raw source code.
Stock photo for illustration only, not from the actual event
The company notes that this tool will initially roll out as a pilot available to a test group of accounts that are at least a year old, before expanding more broadly. Furthermore, developers visiting the GitHub repository will be able to submit updates known as pull requests, which X engineers will consider incorporating into the platform's core algorithm, though not all additions will make the final cut.
"This is the kind of thing that I think people will be fairly shocked that we are releasing,"
X Executive
However, certain systems are intentionally excluded from this release, such as those utilizing Grok to predict whether a post might violate community rules. This restriction protects the platform from bad actors who could exploit the details to bypass regulations and flood the network with spam.
This code release directly addresses ongoing concerns regarding how X's underlying algorithm influences politics, elections, and the spread of misinformation. Before the company was taken private, congressional lawmakers had long accused the social network of leaning politically left and shadowbanning their posts—allegations the platform consistently denied. While the code release boosts transparency in this specific area, it arrives during a period when X has become less forthcoming with overall user metrics, revenue data, and government takedown requests following its transition back to a private entity.
Source: TechCrunch
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment