Towards a Privacy and Security-Aware Framework for Ethical AI: Guiding the Development and Assessment of AI Systems

Privacy and security are foundational to ethical AI: The paper identifies privacy and security as critical dimensions of AI ethics that organizations currently struggle to evaluate, highlighting that these aren't optional add-ons but essential components of responsible AI development.

There's a gap between regulation and practice: While the EU AI Act represents the first comprehensive AI law, organizations lack practical tools to implement privacy and security standards, suggesting that legal frameworks alone are insufficient without accompanying guidance for implementation.

A unified framework is needed across sectors: The research synthesizes literature from 2020-2023 to create a single conceptual framework intended to help organizations, academic institutions, and governments assess AI systems consistently, addressing the fragmentation in how different stakeholders currently approach AI ethics.

Multiple stakeholders must be involved in AI governance: The framework is explicitly designed for diverse actors—not just technologists—including organizations, academic institutions, and governmental bodies, emphasizing that responsible AI requires coordinated effort across sectors rather than isolated efforts.

Ethical decision-making requires systematic guidance: The paper proposes the framework as a guide for ethical decision-making in AI development and utilization, suggesting that organizations need structured tools to navigate privacy and security considerations rather than relying on ad-hoc approaches.