Neither Layer Alone: Epistemic Integrity Requires Hierarchical Joint Design for Long-Running AI Agents

2026-05-31

AI Summary

This work addresses the semantic inconsistency that arises in long-running AI agents due to the independent evolution of model and execution layers, which undermines belief coherence, capability reliability, and goal commitment—referred to as the “interface volatility” problem. To mitigate this, the paper introduces interface contracts as a foundational design principle and proposes a hierarchical four-layer contract framework grounded in Agent Epistemic Integrity (AEI) as an architectural constraint. This framework encompasses goal validity, action prototype sequencing, tool instance selection, and call-level failure discrimination. By co-designing the model and execution layers and integrating structured output generation, cross-layer semantic alignment, and contract-driven training and evaluation, the approach ensures cross-session state consistency even when any single layer is independently updated, thereby establishing a verifiable and maintainable architecture for long-term autonomous agents.

Abstract

Long-running AI agents fail not only when inference fails or tools are underspecified, but when independently evolving model and harness layers change the semantics of belief, capability, and goal commitments across their boundary - a failure class this paper terms Interface Volatility. This paper argues that Agent Epistemic Integrity (AEI) must be treated as a first-class architectural constraint, achievable only through joint model-harness design organized around an explicit interface contract. The central claim is that the model-harness interface contract is the precondition for joint design; its operational form is a four-level hierarchy - goal validity, action-archetype sequencing, tool-instance selection, and invocation-level failure discrimination - that specifies what the boundary must preserve and what structured outputs the model must return for the contract to hold across levels. This reframes long-running agent design away from flat action loops and toward contract-preserving control over persistent state. Evaluation and training should therefore derive from the contract itself, testing whether belief, tool, and goal commitments hold across session boundaries and independent layer upgrades.

Problem

• Interface Volatility
• Agent Epistemic Integrity
• model-harness interface
• long-running AI agents
• semantic consistency

Innovation

• Interface Volatility
• Agent Epistemic Integrity
• Joint Model-Harness Design
• Interface Contract
• Hierarchical Control