Ask HN: Building a ShadowAI detection tool, need inputs from the community
Hello All,
I am building a tool for detecting shadow AI (or Embedded AI). My process involves ingesting logs and classifying them as either shadow AI or not, then returning a CSV.
I want to improve it more and am looking for some input on what else I can add to the dashboard?
What would be meaningful as a data point aside from just identifying whether this tool is Shadow/Embedded AI or not.
Would appreciate any help to make it more meaningful to the end user.
Thank you!
Comments
To add: I can provide information about the data security practices of the tools, including details on data sharing, any identified security vulnerabilities, and their access to sensitive data.