Belief Filtering for Epistemic Control in Linguistic State Space
This addresses AI safety and alignment through architecturally embedded cognitive governance, though it appears incremental as it builds on existing linguistic representation frameworks.
The paper tackles the problem of regulating artificial agents' internal cognitive states by developing belief filtering mechanisms within the Semantic Manifold framework, where belief states are dynamic ensembles of natural language fragments, offering a principled approach to agent regulation that enhances AI safety and alignment.
We examine belief filtering as a mechanism for the epistemic control of artificial agents, focusing on the regulation of internal cognitive states represented as linguistic expressions. This mechanism is developed within the Semantic Manifold framework, where belief states are dynamic, structured ensembles of natural language fragments. Belief filters act as content-aware operations on these fragments across various cognitive transitions. This paper illustrates how the inherent interpretability and modularity of such a linguistically-grounded cognitive architecture directly enable belief filtering, offering a principled approach to agent regulation. The study highlights the potential for enhancing AI safety and alignment through structured interventions in an agent's internal semantic space and points to new directions for architecturally embedded cognitive governance.