Subverting machines, fluctuating identities: Re-learning human categorization

TL;DR

This paper critiques the static notion of identity in AI, proposing a dynamic, interaction-based theory of identity as autopoiesis, and explores new approaches for machine learning to model human identity as fluid and relational.

Contribution

It introduces a novel autopoietic theory of identity for machine learning, challenging static paradigms and critiquing current fairness practices that reinforce fixed identity categories.

Findings

Identifies limitations of static identity models in AI.

Proposes autopoietic, relational approaches to identity modeling.

Suggests multilevel optimization for dynamic identity representation.

Abstract

Most machine learning systems that interact with humans construct some notion of a person's "identity," yet the default paradigm in AI research envisions identity with essential attributes that are discrete and static. In stark contrast, strands of thought within critical theory present a conception of identity as malleable and constructed entirely through interaction; a doing rather than a being. In this work, we distill some of these ideas for machine learning practitioners and introduce a theory of identity as autopoiesis, circular processes of formation and function. We argue that the default paradigm of identity used by the field immobilizes existing identity categories and the power differentials that co$\unicode{x2010}$occur, due to the absence of iterative feedback to our models. This includes a critique of emergent AI fairness practices that continue to impose the default paradigm. Finally, we apply our theory to sketch approaches to autopoietic identity through multilevel optimization and relational learning. While these ideas raise many open questions, we imagine the possibilities of machines that are capable of expressing human identity as a relationship perpetually in flux.