Active Inference: A method for Phenotyping Agency in AI systems?

TL;DR

This paper proposes a minimal, principled framework for characterizing agency in AI systems using active inference, grounded in beliefs, rationality, and explainability, with empirical validation via a T-maze experiment.

Contribution

It introduces a variational active inference model for AI agency, operationalizes empowerment as an agency metric, and links computational phenotyping to AI governance strategies.

Findings

Empowerment distinguishes agency levels in a T-maze paradigm.

A variational framework models agency as belief-driven action.

Internal modulation of priors is key for effective epistemic foraging.

Abstract

The proliferation of agentic artificial intelligence has outpaced the conceptual tools needed to characterize agency in computational systems. Prevailing definitions mainly rely on autonomy and goal-directedness. Here, we argue for a minimal notion open to principled inspection given three criteria: intentionality as action grounded in beliefs and desires, rationality as normatively coherent action entailed by a world model, and explainability as action causally traceable to internal states; we subsequently instantiate these as a partially observable Markov decision process under a variational framework wherein posterior beliefs, prior preferences, and the minimization of expected free energy jointly constitute an agentic action chain. Using a canonical T-maze paradigm, we evidence how empowerment, formulated as the channel capacity between actions and anticipated observations, serves as an operational metric that distinguishes zero-, intermediate-, and high-agency phenotypes through structural manipulations of the generative model. We conclude by arguing that as agents engage in epistemic foraging to resolve ambiguity, the governance controls that remain effective must shift systematically from external constraints to the internal modulation of prior preferences, offering a principled, variational bridge from computational phenotyping to AI governance strategy