The vulnerability arises from the lack of a limit on the number of characters in experiment names within MLflow. By creating or renaming an experiment with a very large number of integers in its name, an attacker can cause the MLflow UI to become unresponsive. This results in a denial of service, preventing users from accessing or managing their experiments. The issue also extends to the artifact_location parameter, which similarly lacks character limits.

An attacker could exploit this vulnerability by using a tool like BurpSuite to create or rename an experiment with a very large number of integers in its name. Once the experiment is created or renamed, refreshing the MLflow UI page would cause it to become unresponsive, effectively denying service to legitimate users.

Users and administrators of MLflow v2.13.2 who rely on the MLflow UI for managing and tracking machine learning experiments are affected by this vulnerability.