Agentic AI: Vision and challenges
Агентный искусственный интеллект: видение и вызовы
2026-08-18
SCID: 54.1/p49egq22
Discuss with AI
Agentic AILarge Language Modelsautonomous operationoutput instabilitysystem integration
Figures from the paper
Abstract (AI)
Agentic AI systems are increasingly viewed as a viable response to the shortcomings of static, rigid, and human-in-the-loop Artificial Intelligence (AI) systems. This is because autonomous operation enables rapid adaptation to dynamic, complex problems with improved time-critical behaviour under real-world constraints. Despite significant progress, current agentic pipelines are still challenged by output instability, scalability gaps, and system integration issues. Addressing these limitations, this article presents a comprehensive conceptual framework unifying core AI functionality with implementation approaches across different system scales, including Agentic AI builds upon Large Language Models (LLMs). Furthermore, the popular applications of Agentic AI and areas for future investigation and open problems are systematically presented.
Key Findings
1
Agentic AI is presented as a viable alternative to static, rigid, and human-in-the-loop AI systems through autonomous operation.
2
Autonomous agentic operation can improve adaptation to dynamic, complex problems and time-critical behavior under real-world constraints.
3
Current agentic AI pipelines remain limited by output instability, scalability gaps, and system integration challenges.
4
The article proposes a conceptual framework unifying core AI functionality with implementation approaches across different system scales, including LLM-based agents.
5
The article systematically reviews popular Agentic AI applications and identifies future research directions and open problems.
Research Object
Agentic AI systems
Research Subject
their autonomous operation, conceptual architecture, applications, limitations, and future research challenges
Publication Details
Publication Date
2026-08-18
Journal
Publisher
ISSN
Cited by
1
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest