The Alignment Gate: Intent And Instruction Guardrails For Agentic AI

Author

• Akash Borigi
• Peggy Lindner, Missouri University of Science and TechnologyFollow
• Alexander Schlager
• Saifullah Shoaib
• Rupendra Lekkala
• Sai Sowjanya Bhamidipati
• Amaury Lendasse, Missouri University of Science and TechnologyFollow

Abstract

This paper proposes an alignment framework for Agentic AI systems, designed to map user intents to corresponding system instructions through interpretable probabilistic associations. The framework introduces a min-median threshold rule to determine whether an instruction is plausibly linked to a given intent, providing a tunable balance between strict and lenient execution criteria. The approach is both lightweight and explainable, enabling clear visualization of alignment scores and transparent control over execution decisions. At this stage, the goal is not to obtain the optimal or final alignment verification mechanism, but rather to assess feasibility and establish a structured foundation for future, more comprehensive alignment frameworks. The method supports modern AI governance by offering a scalable, interpretable path to safer Agentic AI.

Recommended Citation

A. Borigi et al., "The Alignment Gate: Intent And Instruction Guardrails For Agentic AI," Esann 2026 Proceedings 34th European Symposium on Artificial Neural Networks Computational Intelligence and Machine Learning, pp. 357 - 362, European Symposium on Artificial Neural Networks (ESANN), Jan 2026.

The definitive version is available at https://doi.org/10.14428/esann/2026.ES2026-172

Department(s)

Engineering Management and Systems Engineering

International Standard Book Number (ISBN)

978-287587096-4

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2026 European Symposium on Artificial Neural Networks (ESANN), All rights reserved.

Publication Date

01 Jan 2026