AI Risk Assessment

Use this tool to evaluate a potential AI project against a 4-step risk assessment framework. Please fill in the project details and score each question from 1 to 4 to proceed.

0%

STEP 1: PRIORITIZE - Should we use GenAI for this?

● Business Case

1. Will this save significant time or cost?
1 (Not at all) 4 (Highly)
2. Does this improve customer/employee experience?
1 (Not at all) 4 (Highly)
3. Is this a repetitive task that humans find tedious?
1 (Not at all) 4 (Highly)
4. Do we have success metrics defined?
1 (Not at all) 4 (Highly)

● Risk & Readiness

5. Is the data we'd use clean and representative?
1 (Not at all) 4 (Highly)
6. Can we start with a pilot/limited scope?
1 (Not at all) 4 (Highly)
7. Do we have someone to oversee this project?
1 (Not at all) 4 (Highly)
8. Is this low-stakes if it goes wrong initially?
1 (Not at all) 4 (Highly)

STEP 2: APPLY - How do we implement safely?

● Data & Privacy

9. Are we using only data we're allowed to use?
1 (Not at all) 4 (Highly)
10. Have we removed sensitive personal information?
1 (Not at all) 4 (Highly)
11. Do we know where our data will be stored/processed?
1 (Not at all) 4 (Highly)
12. Can we delete data from the AI system if needed?
1 (Not at all) 4 (Highly)

● Governance & Controls

13. Do we have approval from management and legal?
1 (Not at all) 4 (Highly)
14. Are there humans reviewing AI outputs before they're used?
1 (Not at all) 4 (Highly)
15. Have we trained users on limitations and proper use?
1 (Not at all) 4 (Highly)
16. Is there a clear escalation path for problems?
1 (Not at all) 4 (Highly)

STEP 3: MEASURE - Is it working as expected?

● Performance Tracking

17. Are we hitting our success metrics?
1 (Not at all) 4 (Highly)
18. Is the AI output quality consistent over time?
1 (Not at all) 4 (Highly)
19. Are users actually adopting and using the tool?
1 (Not at all) 4 (Highly)
20. Are we seeing the expected time/cost savings?
1 (Not at all) 4 (Highly)

● Risk Monitoring

21. Are we checking the outputs for bias or unfairness?
1 (Not at all) 4 (Highly)
22. Have there been any errors or inappropriate responses?
Note: For this question, 4 = No errors (best), 1 = Many errors (worst)
1 (Many errors) 4 (No errors)
23. Are we staying within budget/resource limits?
1 (Not at all) 4 (Highly)
24. Is the AI making decisions we can explain if asked?
1 (Not at all) 4 (Highly)

STEP 4: ASSESS - Should we continue/expand?

● Value & Impact

25. Has this delivered measurable business value?
1 (Not at all) 4 (Highly)
26. Are the benefits worth the costs and risks?
1 (Not at all) 4 (Highly)
27. Would we recommend this to other teams?
1 (Not at all) 4 (Highly)
28. What have we learned we can apply elsewhere?
1 (Not at all) 4 (Highly)

● Ongoing Viability

29. Can we maintain this long term?
1 (Not at all) 4 (Highly)
30. Are there any regulatory or policy changes we should consider?
1 (Not at all) 4 (Highly)
31. Should we expand, modify or discontinue this use case?
1 (Discontinue) 4 (Expand)
32. What's our plan for the next review cycle?
1 (No plan) 4 (Detailed plan)

Assessment Results

Your Project Score:

Step-by-Step Scores

Project Classification