You hold health records, housing status, immigration information, abuse histories, children's data. Your donors trust you with their generosity. When you adopt AI tools without understanding what happens to that data, you risk betraying both. For organizations working in advocacy, human rights, immigration, or domestic violence services, a data exposure could endanger lives.
That data has just been sent to a third party's servers. Depending on the tool and plan tier, it may be used for model training, stored indefinitely, or exposed in a data breach. This is happening daily at organizations across the country. The solution: a clear policy that no donor names, contact information, constituent records, case notes, or financial data enters any AI tool without leadership approval and a verified enterprise-tier account.
- No donor names, contact info, or constituent data into AI tools without leadership approval. High
- Audit biometric data collection (time clocks, security cameras, photo databases). High
- Brief your board on AI governance (even 15 minutes next meeting). High
- Be LLM-agnostic. Diversify AI vendors. Medium
Next Steps
Your mission depends on trust. Keep building it.