Résumé : Society's increasing dependence on Artificial Intelligence (AI) and AI-enabled systems require a more practical approach from software engineering (SE) executives in middle and higher-level management to improve their involvement in implementing AI ethics by making ethical requirements part of their management practices. However, research indicates that most work on implementing ethical requirements in SE management primarily focuses on technical development, with scarce findings for middle and higher-level management. We investigate this by interviewing ten Finnish SE executives in middle and higher-level management to examine how they consider and implement ethical requirements. We use ethical requirements from the European Union (EU) Trustworthy Ethics guidelines for Trustworthy AI as our reference for ethical requirements and an Agile portfolio management framework to analyze implementation. Our findings reveal a general consideration of privacy and data governance ethical requirements as legal requirements with no other consideration for ethical requirements identified. The findings also show practicable consideration of ethical requirements as technical robustness and safety for implementation as risk requirements and societal and environmental well-being for implementation as sustainability requirements. We examine a practical approach to implementing ethical requirements using the ethical risk requirements stack employing the Agile portfolio management framework.