AI and Matching Risk Assessment

1. Use Case

Describe whether WorkTwins is used for:

• informal collaboration;
• team matching;
• employee matching;
• candidate matching;
• company hiring support;
• task allocation;
• performance analysis.

2. Output

Describe outputs:

• compatibility score;
• match ranking;
• “matched because” explanation;
• WorkFootPrint summary;
• collaboration suggestion.

3. Decision Impact

Could outputs affect:

• hiring;
• promotion;
• compensation;
• discipline;
• task assignment;
• access to opportunities;
• social/professional reputation?

4. Risk Controls

• no sole automated employment decisions;
• human review;
• user notice;
• ability to contest/reprocess;
• bias review;
• data minimization;
• limited explanations;
• no raw context exposure;
• audit logging.

5. Bias and Fairness

Assess whether matching may disadvantage protected groups or workers with different tools, schedules, disabilities, languages, or work styles.

6. Transparency

Confirm users understand:

• what WorkTwins does;
• how matching works;
• what companies can see;
• what can be excluded or controlled.

7. Conclusion

Risk level: Required mitigations: Approval: Review date: