Enhanced well-being assessment as basis for the practical implementation of ethical and rights-based normative principles for AI

Artificial Intelligence (AI) has an increasing impact on all areas of\npeople's livelihoods. A detailed look at existing interdisciplinary and\ntransdisciplinary metrics frameworks could bring new insights and enable\npractitioners to navigate the challenge of understanding and assessing the\nimpact of Autonomous and Intelligent Systems (A/IS). There has been emerging\nconsensus on fundamental ethical and rights-based AI principles proposed by\nscholars, governments, civil rights organizations, and technology companies. In\norder to move from principles to real-world implementation, we adopt a lens\nmotivated by regulatory impact assessments and the well-being movement in\npublic policy. Similar to public policy interventions, outcomes of AI systems\nimplementation may have far-reaching complex impacts. In public policy,\nindicators are only part of a broader toolbox, as metrics inherently lead to\ngaming and dissolution of incentives and objectives. Similarly, in the case of\nA/IS, there's a need for a larger toolbox that allows for the iterative\nassessment of identified impacts, inclusion of new impacts in the analysis, and\nidentification of emerging trade-offs. In this paper, we propose the practical\napplication of an enhanced well-being impact assessment framework for A/IS that\ncould be employed to address ethical and rights-based normative principles in\nAI. This process could enable a human-centered algorithmically-supported\napproach to the understanding of the impacts of AI systems. Finally, we propose\na new testing infrastructure which would allow for governments, civil rights\norganizations, and others, to engage in cooperating with A/IS developers\ntowards implementation of enhanced well-being impact assessments.\n