Red-teaming university policy

Here’s an experiment.

1. Upload the university assessment policy to genAI (I’d use Claude, for it’s context window).
2. Ask Claude to identify all the ways an adversarial student could find loopholes in the policy.
3. Ask Claude for modifications that might help close the loopholes.

Do the same for your recruitment policy, and so on.

In a variation of the theme, upload ALL your policies and ask genAI to find all the places where they contradict each other. Or where they seem overly restrictive. Or punitive. Or [insert concern here].