ENABLIENT RESPONSIBLE AI CHECKLIST
=================================

Use this checklist before launching an AI-enabled workflow.

1. Workflow purpose
- What business process will AI support?
- What outcome should improve?
- How will time savings or quality improvement be measured?

2. Data review
- What data will the workflow use?
- Does it include client, donor, patient, employee, financial, regulated, or confidential information?
- Where is the data stored?
- Does the AI tool retain prompts, files, or outputs?

3. Access control
- Who can use the workflow?
- Who can see inputs and outputs?
- Are permissions limited to people who need access?

4. Human oversight
- What outputs require human approval?
- Who is accountable for final decisions?
- What should users do if the AI output is wrong or uncertain?

5. Tool approval
- Is the tool approved for this data type and workflow?
- Does the tool fit the budget, security needs, and maintainability requirements?
- Is there a lower-risk or simpler alternative?

6. Monitoring
- What errors, exceptions, or unusual outputs should be tracked?
- How often will the workflow be reviewed?
- Who owns improvements after launch?

7. Staff readiness
- Have users been trained on allowed and prohibited uses?
- Do users understand sensitive-data rules?
- Do users know when to escalate to a person?

Recommended next step:
Schedule a trust and workflow review before scaling AI across more processes.