Institution profile

CEA Saclay

Academic institutioneurope · fr
Official website
Research library2linked papers
Opportunities0open roles
Selected work

Representative Papers

Ethics Readiness of Artificial Intelligence: A Practical Evaluation Method

Dec 10, 2025

This study addresses the challenge of operationalizing ethical principles in AI system design. Methodologically, it introduces Ethical Readiness Levels (ERLs)—a four-tier iterative framework that translates abstract ethical principles into contextualized design prompts, checklist items, and governance mechanisms. It innovates through dynamic tree-structured questionnaires, a context-sensitive metric system, and a multi-level scoring scheme enabling domain- and technology-adaptive assessment; additionally, it establishes an interdisciplinary workflow that formalizes structured collaboration between ethics experts and engineering teams. Evaluated in two real-world applications—face sketch generation and collaborative industrial robotics—the ERL framework significantly enhances the visibility, traceability, and actionable impact of ethical practice throughout the development lifecycle. Results demonstrate a paradigm shift from technocentric solutionism toward ethics-by-design in AI development.

0 citationsRead paper
Recent publications

Latest Papers

Ethics Readiness of Artificial Intelligence: A Practical Evaluation Method

Dec 10, 2025

This study addresses the challenge of operationalizing ethical principles in AI system design. Methodologically, it introduces Ethical Readiness Levels (ERLs)—a four-tier iterative framework that translates abstract ethical principles into contextualized design prompts, checklist items, and governance mechanisms. It innovates through dynamic tree-structured questionnaires, a context-sensitive metric system, and a multi-level scoring scheme enabling domain- and technology-adaptive assessment; additionally, it establishes an interdisciplinary workflow that formalizes structured collaboration between ethics experts and engineering teams. Evaluated in two real-world applications—face sketch generation and collaborative industrial robotics—the ERL framework significantly enhances the visibility, traceability, and actionable impact of ethical practice throughout the development lifecycle. Results demonstrate a paradigm shift from technocentric solutionism toward ethics-by-design in AI development.

0 citationsRead paper