NSF Award
2412398
The proliferation of Artificial Intelligence (AI)-enabled technologies is transforming the world. AI has brought promise and peril to nearly every aspect of human life. Nurturing AI engineers who can design AI systems responsibly has become a key objective from business and policy perspectives. For engineering educators committed to supporting students to become competent and responsible engineers, thoughtful consideration of how to effectively incorporate ethical principles into the classroom is paramount. However, translating existing research-based AI ethics tools, such as ethics principles, into engineering practice and education is not easy. Ethical principles are often too abstract to be integrated into design practices, as they do not provide clear explanations or procedural details on how they can be integrated into technical practice. Therefore, scholars have recently called for a transition in AI ethics research toward a translational approach, which focuses on ensuring that AI practitioners can implement AI ethics tools such as ethical principles into actual AI systems designs. However, there is a lack of empirical evidence for the extent to which engineers integrate these ethical principles into their daily practices of AI developers, which generates challenges for engineering educators to incorporate these resources into their own teaching and educational research.<br/><br/>This research aims to advance the fundamental knowledge of the formation of future AI engineers poised to innovate responsibly and construct advanced engineering systems empowered by AI technologies. It will investigate a practical competency critical for socially responsible AI engineering. This includes translational ethical competency, or the ability to translate general ethical principles and values into specific decisions in AI engineering practices in the day-to-day work of AI engineers. To achieve this, researchers from diverse backgrounds including computer science, engineering education, and the ethics of engineering will investigate how to define and cultivate a “translational competency” that allows engineers to translate general AI ethical principles into engineering practices in AI system designs. This project will address the following key research questions: (1) What processes do AI engineers follow to translate ethical principles into engineering practices? (2) What competencies are critical for translating ethical principles into engineering practices (translational competencies)? and (3) How do AI engineers develop the translational competencies for responsible AI engineering? Answering these research questions is critical to ensure future engineering professionals have sufficient learning opportunities to integrate ethical values into real-world engineering problems. The project aligns with the Research in the Formation of Engineers initiative goal of advancing understanding of the educational and professional formation of engineers. Specifically, the research seeks to improve the development of professional and technical skills in postsecondary education through a more practical approach to engineering ethics and to improve the transition from education settings to the workforce for engineers seeking to use what they learn in ethics courses in their future work. Additionally, this project seeks to improve the relationship between engineering and the public by equipping engineers to more effectively handle real-world challenges related to the ethical deployment of AI, including bias, privacy, and autonomy. This research is thus necessary to improve ethics education for engineers, and to help bridge the gap between classroom and job site.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
