A Theory of LLM Information Susceptibility

TL;DR

This paper develops a theoretical framework to understand the limits of large language models in improving agentic systems, highlighting conditions where nested architectures enable better self-improvement.

Contribution

It introduces a new theory of LLM information susceptibility, generalizes it to multi-channel architectures, and empirically validates the conditions for effective LLM intervention.

Findings

Nested, co-scaling architectures enable response channels unavailable to fixed configurations.

The theory provides predictive constraints for AI system design.

Empirical validation across diverse domains supports the susceptibility hypothesis.

Abstract

Large language models (LLMs) are increasingly deployed as optimization modules in agentic systems, yet the fundamental limits of such LLM-mediated improvement remain poorly understood. Here we propose a theory of LLM information susceptibility, centred on the hypothesis that when computational resources are sufficiently large, the intervention of a fixed LLM does not increase the performance susceptibility of a strategy set with respect to budget. We develop a multi-variable utility-function framework that generalizes this hypothesis to architectures with multiple co-varying budget channels, and discuss the conditions under which co-scaling can exceed the susceptibility bound. We validate the theory empirically across structurally diverse domains and model scales spanning an order of magnitude, and show that nested, co-scaling architectures open response channels unavailable to fixed configurations. These results clarify when LLM intervention helps and when it does not, demonstrating that tools from statistical physics can provide predictive constraints for the design of AI systems. If the susceptibility hypothesis holds generally, the theory suggests that nested architectures may be a necessary structural condition for open-ended agentic self-improvement.