AI Values¶
SUM Values: Respect, Connect, Care, and Protect¶
In keeping with this hazards-responsive approach, the SUM Values incorporate conceptual elements from both bioethics and human rights discourse, but they do so with an eye to applying the most critical of these elements to the specific social and ethical problems raised by the potential misuse, abuse, poor design, or harmful unintended consequences of AI systems.
The SUM Values are values that support, underwrite, and motivate a responsible innovation ecosystem. Their role is to provide an accessible framework of ethical criteria for considering, assessing, and deliberating on the ethical permissibility of a prospective AI project and its ethical impacts. They are meant to be utilised as guiding values throughout the innovation lifecycle: from the preliminary steps of project evaluation, planning, and problem formulation, through processes of design, development, and testing, to the stages of implementation and reassessment. Adopting common values from the outset enables reciprocally respectful, sincere, and open dialogue about ethical challenges by helping to create a shared vocabulary for informed dialogue and impact assessment. Such a common starting point also facilitates discussion and deliberation about how to balance values when they come into conflict.
The SUM Values encompass a range of values that are distilled in the following four (corresponding ethical concerns are indicated in the circle):
