Practice question
A court proposes an AI tool trained on incomplete data that systematically disadvantages a vulnerable group. The proposal calls the tool neutral solely because it is automated and requires officials to accept every output. Which response best follows the Framework?
- Automation proves neutrality
- Incomplete training data cannot cause bias
- Vulnerable groups are outside the fairness principle
- The proposal conflicts with fairness and human-oversight requirements; automation neither guarantees neutrality nor justifies compulsory unreviewed adoption
- The vendor alone decides whether constitutional rights matter
Legal references
- A.M. No. 25-11-28-SC, February 18, 2026, AI Governance Framework, Part II (human-centered use, transparency, accountability and fairness): https://elibrary.judiciary.gov.ph/thebookshelf/showdocs/11/101138
- 2025 CJCA, Canon VI, Sections 5–6: https://elibrary.judiciary.gov.ph/thebookshelf/showdocs/11/100836
Answer this question in flashcards to review your response and continue practicing.