‘The names have changed, but the game’s the same’: artificial intelligence and racial policy in the USA

Abstract

Like the Operations Research models used to justify the ethnic cleansing of minority voting blocs in 1970’s New York City, AI ‘risk assessment’ systems for individuals will be used to reinforce longstanding power relations between ethnic groups within the USA. From the perspective of African–Americans and their abolitionist allies, the central problem with AI risk assessment does not involve ‘corrective’ stabilization of an inadvertently unstable system. On the contrary, that system’s de-facto—if sometimes camouflaged—purpose is enforcing the stability of historic patterns of racial oppression, constitutional formalities notwithstanding. AI, like ‘OR’ before it, becomes, then, simply another tactic in a persistent strategy aimed at reinforcing a stable cultural trajectory with roots deep in human slavery. To the archetypic question ‘what is to be done?’ is the archetypic answer: build countervailing power.