Whether your data is fit for the use you have in mind. It breaks down into a few practical questions: is it complete, is it accurate, is it consistent, is it current, and is it actually about what you think it is about. Most AI projects that fail on data fail on the last one.
Technical view
Standard dimensions are completeness, accuracy, consistency, timeliness, validity and uniqueness, each measurable as a rule-based check that can be automated in a pipeline. Tools express these as assertions that fail a build. The dimension that resists automation is fitness for purpose, which requires knowing how the data was generated and what it omits, and this is where documentation matters more than tooling.
Related
Part of AI Terminology Index, section: Data.