Table 4.

Ethical models and frameworks for AI development.

Model	Description	Key features	Reference
Embedded ethics (EE)	A collaborative approach where ethicists and developers work together throughout the AI development process.	Continuous integration of ethical considerations Addresses ethical issues from planning to implementation	McLennan et al. (2022)
Embedded ethics for responsible AI systems (EE-RAIS)	Focuses on ethical, legal, and social values in AI systems, particularly in disaster management.	Four platforms: educational, cross-functional, developmental, algorithmic Metrics: ethical intelligence, legal intelligence, social–emotional competency, artificial wisdom	Afroogh et al. (2023)
Human-in-the-loop (HITL)	Incorporates human oversight at various stages of AI development to ensure ethical outputs.	Regular review of AI outputs Balances commercial goals with social impacts	Middleton et al. (2022)
Responsible intelligent systems	Emphasizes moral responsibility and accountability in the use of intelligent systems.	Concept of ecosystems in AI Higher-level responsibility or ‘meta-responsibility’	Stahl (2023)
Socio-technical approaches	Integrates social and technical perspectives to address ethical issues in AI development.	Collaboration across disciplines Focus on societal implications of AI technologies	McLennan et al. (2022); McLennan et al. (2024)