Can AI be deployed in ways that enhance equality, or will AI systems exacerbate existing structural inequalities and create new inequities? The Carnegie AI & Equality Initiative (AIEI) is directed at understanding the innumerable ways in which artificial intelligence impacts equality for better or for worse.
Many technological innovations are being proposed as means to ameliorate inequality. Whether such fixes address sensitive trade-offs remains unclear. There will be unanticipated consequences. However, undesirable societal impacts and risks that are perceived in advance can usually be mitigated, presuming there is the will to do so.
Structural inequality is the result of a broad array of political, economic, social and cultural factors. The socio-technical systems that are the result of introducing innovations into this mix have become increasingly destabilizing. The sheer ubiquity and speed by which AI based systems are permeating our lives is disruptive of countless industries and institutions. Growing monopolies of proprietary data have and continue to rapidly empower digital elites and new digital alliances. And yet the understanding of exactly how social and technical systems interact and how to govern them globally, regionally or locally lags far behind.
The Carnegie AIEI proposes to unpack this difficult and highly transdisciplinary terrain. The project will map the field, illuminate topics requiring further research, build a diverse community of experts to share their insights, and generate educational materials. We expect to propose mechanisms that could be implemented to ensure the benefits of AI for all people, with a particular focus on vulnerable communities at risk. Our priority is to ensure that AI is developed and deployed in a just, responsible and inclusive manner.