Beyond Politeness and Indirectness: A Unified Framework for Autonomous, Unidirectional, and Bidirectional Direct Expression in Generative AI–Human Communication

Generative artificial intelligence (AI) has transformed human–computer communication by enabling increasingly sophisticated natural-language interactions. However, communication systems frequently introduce unnecessary linguistic modifications, including excessive politeness, indirectness, hedging, introductory statements, social cushioning, redundant explanations, and other forms of expressive elaboration. Although these strategies can support interpersonal relationships and reduce potential face threats, their unnecessary application may obscure the speaker's actual position, increase communication overhead, and weaken the immediacy of information exchange. Existing research has investigated AI politeness, communication styles, linguistic directness, and human–AI pragmatic alignment, but a comprehensive model of independently controlled direct expression by both human users and generative AI remains insufficiently established. This paper introduces Autonomous Direct Expression and Directionality Framework (ADEDF), a proposed theoretical framework for representing direct expression as an independently selectable communication capability rather than a fixed personality characteristic or an exclusively AI-controlled stylistic preference. The framework distinguishes six communication configurations: bidirectional direct expression, AI-to-human unidirectional direct expression, human-to-AI unidirectional direct expression, conditional AI direct expression, conditional human direct expression, and independently selectable direct expression. It further conceptualizes directness through seven dimensions: expression directness, linguistic modification, opinion transparency, expression autonomy, communication directionality, contextual adaptability, and directness boundaries. A formal representation is developed to distinguish the expression policies of the two communication participants, while a set of research propositions establishes relationships among directness, communication efficiency, interpretability, perceived authenticity, interpersonal respect, and contextual appropriateness. The paper also proposes an evaluation methodology for future empirical validation. Rather than equating directness with impoliteness or unrestricted bluntness, ADEDF conceptualizes direct expression as the accurate and sufficiently unmodified communication of intended meaning while preserving truthfulness, respect, uncertainty disclosure, and safety. To the best of our knowledge, this paper presents the first unified framework specifically organized around independently selectable human–AI direct-expression authority and its directional configurations. The proposed framework establishes a foundation for research into communication autonomy, expressive control, and configurable interaction policies in generative AI systems.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24298267
Primary Topic
Language, Discourse, Communication Strategies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Beyond Politeness and Indirectness: A Unified Framework for Autonomous, Unidirectional, and Bidirectional Direct Expression in Generative AI–Human Communication

Jincheng Zhang
10 links
Zenodo (CERN European Organization for Nuclear Research)
Language, Discourse, Communication Strategies
preprint

Beyond Politeness and Indirectness: A Unified Framework for Autonomous, Unidirectional, and Bidirectional Direct Expression in Generative AI–Human Communication

Jincheng Zhang
preprint en

Abstract

Generative artificial intelligence (AI) has transformed human–computer communication by enabling increasingly sophisticated natural-language interactions. However, communication systems frequently introduce unnecessary linguistic modifications, including excessive politeness, indirectness, hedging, introductory statements, social cushioning, redundant explanations, and other forms of expressive elaboration. Although these strategies can support interpersonal relationships and reduce potential face threats, their unnecessary application may obscure the speaker's actual position, increase communication overhead, and weaken the immediacy of information exchange. Existing research has investigated AI politeness, communication styles, linguistic directness, and human–AI pragmatic alignment, but a comprehensive model of independently controlled direct expression by both human users and generative AI remains insufficiently established. This paper introduces Autonomous Direct Expression and Directionality Framework (ADEDF), a proposed theoretical framework for representing direct expression as an independently selectable communication capability rather than a fixed personality characteristic or an exclusively AI-controlled stylistic preference. The framework distinguishes six communication configurations: bidirectional direct expression, AI-to-human unidirectional direct expression, human-to-AI unidirectional direct expression, conditional AI direct expression, conditional human direct expression, and independently selectable direct expression. It further conceptualizes directness through seven dimensions: expression directness, linguistic modification, opinion transparency, expression autonomy, communication directionality, contextual adaptability, and directness boundaries. A formal representation is developed to distinguish the expression policies of the two communication participants, while a set of research propositions establishes relationships among directness, communication efficiency, interpretability, perceived authenticity, interpersonal respect, and contextual appropriateness. The paper also proposes an evaluation methodology for future empirical validation. Rather than equating directness with impoliteness or unrestricted bluntness, ADEDF conceptualizes direct expression as the accurate and sufficiently unmodified communication of intended meaning while preserving truthfulness, respect, uncertainty disclosure, and safety. To the best of our knowledge, this paper presents the first unified framework specifically organized around independently selectable human–AI direct-expression authority and its directional configurations. The proposed framework establishes a foundation for research into communication autonomy, expressive control, and configurable interaction policies in generative AI systems.

Zenodo (CERN European Organization for Nuclear Research)
Language, Discourse, Communication Strategies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon