Data Scans and AI for Data Catalogs/Fabrics (Dataplex & GPT) 

It's great to hear that you found the video interesting! For situations where metadata is not explicitly generated by applications or data transformers, but there is contextual information available from reporting or visualization tools, yes, you can marry the disparate metadata sources as long as the metadata is exposed in those external tools through some sort of interface or API. Dataplex is agnostic to the source and structure of metadata, so you'd just need some sort of metadata ETL pipeline to extract and then load that metadata (possibly with transformations and NLP/GenAI/etc.). I don't believe there are any tools that do this in a turnkey way - you're probably writing Python, optionally through some pipelining framework (e.g., PySpark or Beam).