AI 法律責任歸屬模糊
Operation & Monitoring
Risk Description
When AI systems cause harm, legal liability attribution among developers, deployers, operators, and users remains ambiguous. Current legal frameworks mostly based on direct human causal relationships are difficult to apply to AI autonomous decision scenarios. Particularly in cases where damage results from humans adopting AI recommendations, determining "AI contribution degree" lacks clear standards.
Framework Mappings
iso 23894R8
tw principle問責
tw risk type社會系統性影響（Societal Systemic）
Controls English translation pending
- 責任歸屬框架
- 保險機制
Implementation Steps English translation pending
- 建立組織內部的 AI 決策責任歸屬框架
- 在 AI 系統部署合約中明確各方的責任範圍
- 評估並購買 AI 專用保險覆蓋潛在責任
- 追蹤 AI 相關法律案例發展，及時調整責任策略