> A difference with data is that there are always multiple people involved before code is merged while data can be extracted by a single individual who has acce

> A difference with data is that there are always multiple people involved before code is merged while data can be extracted by a single individual who has access. It sounds like you just need to harden your production perimeter. Jump boxes with two-man-rule access and terminal logging. Apply the same practices to data as you do code. > Discriminating on origin is likely illegal. You should ask your legal folks about the national security exceptions of Title VII. It sounds like your customer requirements are pushing you in that direction anyway.