Unsocial Intelligence: a Pluralistic, Democratic, and Participatory Investigation of AGI Discourse

Dreams of machines that rival human intelligence have shaped the field of AI since its inception. Yet there remains no agreed-upon conception of what human-level AI or artificial general intelligence (AGI) means. We investigate key social, political, and ethical assumptions made by influential conceptions of AGI and human-level AI. We then draw on feminist, STS, and social science scholarship on the political and social character of intelligence in both humans and machines to defend a pluralistic, democratic, and participatory conception of the topic. We argue that framing AGI or human-level AI as a technical or value-neutral topic leads to political, ethical, and epistemic harm. AGI should not be developed without explicit attention to the values they encode, the people they include or exclude, and a view toward epistemic justice.