From Generative Models to Autonomous Agents: A Multidimensional Taxonomy of Generative AI Architectures and Operational Autonomy

The rapid evolution of Generative Artificial Intelligence (GenAI) has transformed the way individuals and organizations interact with computational systems. Initially dominated by foundation models and conversational interfaces, the GenAI landscape increasingly incorporates integrated AI assistants and autonomous agents capable of planning, executing, and monitoring complex tasks. However, the growing diversity of these systems has created conceptual ambiguity regarding their architectural characteristics, functional differences, and degrees of operational autonomy. Existing classifications frequently emphasize individual technological components, such as large language models (LLMs), memory mechanisms, planning capabilities, and tool integration, without sufficiently distinguishing architectural integration from operational autonomy. This paper proposes a multidimensional taxonomy of GenAI architectures to address this conceptual ambiguity. The framework distinguishes four functional architectural configurations: (1) foundation models, (2) conversational interfaces, (3) integrated AI assistants, and (4) AI agents. It further conceptualizes operational autonomy as an independent dimension, comprising four levels: reactive, task-oriented, goal-oriented, and adaptive autonomy. The framework also identifies the supporting technological infrastructure required to enable increasingly sophisticated GenAI systems, including orchestration frameworks, memory mechanisms, vector databases, external tools, execution environments, and governance mechanisms. By separating architectural configuration, operational autonomy, and supporting infrastructure, the proposed taxonomy provides a more comprehensive basis for understanding and comparing GenAI systems. The paper contributes to the literature by offering a conceptual classification framework that can support future research into AI system development, organizational adoption, human-AI collaboration, and the implications of agentic AI for management and information systems.

Authors

Publication Details

Published
2026-11-18
DOI
https://doi.org/10.21428/b3658bca.b19d7837
Primary Topic
Artificial Intelligence Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

From Generative Models to Autonomous Agents: A Multidimensional Taxonomy of Generative AI Architectures and Operational Autonomy

Carlos José de Castro Costa
3 links
Artificial Intelligence Applications
article

From Generative Models to Autonomous Agents: A Multidimensional Taxonomy of Generative AI Architectures and Operational Autonomy

Carlos José de Castro Costa
article en

Abstract

The rapid evolution of Generative Artificial Intelligence (GenAI) has transformed the way individuals and organizations interact with computational systems. Initially dominated by foundation models and conversational interfaces, the GenAI landscape increasingly incorporates integrated AI assistants and autonomous agents capable of planning, executing, and monitoring complex tasks. However, the growing diversity of these systems has created conceptual ambiguity regarding their architectural characteristics, functional differences, and degrees of operational autonomy. Existing classifications frequently emphasize individual technological components, such as large language models (LLMs), memory mechanisms, planning capabilities, and tool integration, without sufficiently distinguishing architectural integration from operational autonomy. This paper proposes a multidimensional taxonomy of GenAI architectures to address this conceptual ambiguity. The framework distinguishes four functional architectural configurations: (1) foundation models, (2) conversational interfaces, (3) integrated AI assistants, and (4) AI agents. It further conceptualizes operational autonomy as an independent dimension, comprising four levels: reactive, task-oriented, goal-oriented, and adaptive autonomy. The framework also identifies the supporting technological infrastructure required to enable increasingly sophisticated GenAI systems, including orchestration frameworks, memory mechanisms, vector databases, external tools, execution environments, and governance mechanisms. By separating architectural configuration, operational autonomy, and supporting infrastructure, the proposed taxonomy provides a more comprehensive basis for understanding and comparing GenAI systems. The paper contributes to the literature by offering a conceptual classification framework that can support future research into AI system development, organizational adoption, human-AI collaboration, and the implications of agentic AI for management and information systems.

Artificial Intelligence Applications
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