Agentic AI and Autonomous Systems
Агентный искусственный интеллект и автономные системы
2026-06-24
SCID: 54.1/k7hd4n85
Discuss with AI
Agentic artificial intelligenceAutonomous systemsLarge language model-powered agentsMulti-agent coordinationRetrieval-Augmented Generation
Figures from the paper
Abstract (AI)
This chapter examines the rapidly evolving paradigm of Agentic Artificial Intelligence and Autonomous Systems, investigating how large language model-powered agents transition from passive query-response tools to active, goal-directed entities capable of multi-step reasoning, tool use, memory management, and self-correction. This chapter per the authors presents a formal taxonomy of agent architectures alongside analysis of enabling frameworks including ReAct, Reflexion, LangGraph, AutoGen, and CrewAI. Core topics include multi agent coordination, Retrieval-Augmented Generation, the Model Context Protocol, neuromorphic computing, and multimodal perception for embodied AI. Case studies spanning autonomous software engineering, clinical decision support, financial trading, and self-driving vehicles are evaluated. The chapter concludes with safety, alignment, governance, and quantum computing trajectories that will amplify agentic AI capabilities,contributing an engineering-first framework essential for researchers, practitioners, and policymakers navigating the Quantum AI era.
Key Findings
1
Agentic AI systems advance beyond passive question answering by supporting goal-directed, multi-step reasoning, tool use, memory management, and self-correction.
2
Case studies examine applications of autonomous agents in software engineering, clinical decision support, financial trading, and self-driving vehicles.
3
Multi-agent coordination, Retrieval-Augmented Generation, the Model Context Protocol, neuromorphic computing, and multimodal perception are identified as key enabling technologies.
4
The chapter emphasizes that safety, alignment, governance, and emerging quantum-computing trajectories are central to the future development of agentic AI.
5
The chapter presents a formal taxonomy of agent architectures and analyzes frameworks including ReAct, Reflexion, LangGraph, AutoGen, and CrewAI.
Research Object
large language model-powered agents and autonomous systems
Research Subject
their transition from passive query-response tools to active, goal-directed entities, including multi-step reasoning, tool use, memory management, self-correction, coordination, perception, and safety
Publication Details
Publication Date
2026-06-24
Journal
Publisher
ISSN
Cited by
2
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
Causality: Models, Reasoning, and Inference2009
Quantum Computing in the NISQ era and beyond2018
Generative Agents: Interactive Simulacra of Human Behavior2023
Artificial Intelligence Risk Management Framework (AI RMF 1.0)2023
Machine learning & artificial intelligence in the quantum domain: a review of recent progress2018
Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection2023