Prediction and Empowerment: A Theory of Agency through Bridge Interfaces

2026-05-07

AI Summary

This work formalizes agency in partially observable deterministic environments and clarifies the interplay among prediction, compression, and empowerment. By constructing an interface model that bridges perception and action, it attributes environmental uncertainty to initial conditions, fixed dynamics, and exogenous noise within a deterministic POMDP framework. The study shows that prediction can be achieved either through latent state identification or overwrite control, demonstrating that high empowerment alone does not guarantee predictive capability, and distinguishes task-relevant controllability from interference suppression. The approach integrates prior-informed latent microstate modeling, many-to-one observation coarse-graining, action-conditional compression, and bit-string specialization under conserved information budgets. Theoretically, under refinable interfaces and sufficient memory, observation compression reduces posterior uncertainty over latent states, yielding alignment principles for modern AI design.

Abstract

We study agency under partial observability in deterministic physical or simulated worlds, where apparent randomness arises from uncertainty over initial conditions, fixed law bits, and unrolled exogenous noise. We model sensing and actuation as bridge interfaces split between agent-controlled parameters and environment-controlled channel state, inducing a deterministic POMDP through a prior over latent microstates and many-to-one observation coarsening. Within this framework, we prove a separation between prediction, compression, and empowerment. Perfect prediction can be achieved either by identifying the hidden quotient relevant to the target family or by overwrite control that makes the future target action-determined; high empowerment alone is insufficient. Under refinable interfaces and sufficient memory, action-conditioned observation-compression progress reduces posterior uncertainty about the latent quotient, and when refinement requires steering world-side channel conditions, this creates target-conditioned interface empowerment. A bit-string specialization with a conserved information budget makes the resulting tradeoff explicit: prediction by identification requires internal capacity at least the relevant latent entropy, whereas overwrite control requires terminal action capacity over the controlled quotient. For modern AI agents, the results suggest a design principle rather than a theorem of inevitability: objectives should distinguish hidden-state identification, interface refinement, task-relevant controllability, and mere overwrite or distractor control. Human--AI alignment is partly an interface-design problem, where the relevant bridge is between human intent, agent internal state, external tools, and world-side channel conditions. This is a working draft: feedback and criticism is most welcome.

Problem

• agency
• partial observability
• prediction
• empowerment
• deterministic POMDP

Innovation

• bridge interfaces
• empowerment
• prediction-compression separation
• latent quotient identification
• overwrite control