Abstract

The rise of AI should not be read as a neutral technological advance but as the latest stage in a long struggle between capital’s drives towards automation, labour’s resistance to exploitation, and society’s quest for autonomy. To situate AI within this history, past theories of automation must be revisited and expanded. A theory of automation, however, is not a study of the ‘impact’ of technology on society, but a causal model to explain how automation develops and why it takes one form rather than another. Such theories can be divided into four groups: value theories, labour theories, metric theories, and standpoint theories of automation. Despite abundant literature, an organic theory of automation to address the predicament of AI and the trade of labour, knowledge, and capital is still lacking.