Unique Presentation Identifier:

25

Program Type

Graduate

Faculty Advisor

Dr. Robin Ghosh, Dr. Abdus Salam Siddique

Document Type

Poster

Location

Face-to-face

Start Date

9-4-2026 1:00 PM

End Date

9-4-2026 3:00 PM

Abstract

In the past several years, the world has managed to transition away from simple automation to independent AI systems. Agentic AI is an agent that can work independently, carrying out all essential plans and implementations without any kind of supervision from a human being. This review has tried to demonstrate the transformative impact that Agentic AI brings to contemporary models of intelligence by means of synthesis of perception, reasoning, and goal. We utilized the phrases Agentic AI, autonomous AI, multi agent systems as keywords in Google Scholar, ScienceDirect, arXiv, and other digital libraries. We have used these 38 main papers published between 2023 and present as part of search process. This paper has taken a deep, close approach to Agentic AI systems, thereby discussing their technical foundations, challenges, applications, and societal impacts. In this paper, we discuss the different aspects of Agentic AI, including the technical side, challenges, applications, and impacts. In this study we found Agentic AI systems are capable of a 34.2 % reduction in the time of task execution, a 7.7 % increase in accuracy, and a 13.6 % improvement in resource utilization. In this survey, we investigate the performance and benchmarking of model architectures, security, privacy, and human-in-the-loop integration; and analyze frameworks for efficient human-AI collaboration. This survey draws on ideas from recent publications, equipping researchers with a holistic view of the state, challenges, and directions for future research necessary to keep Agentic AI safe, aligned, and beneficial to society.

Share

COinS
 
Apr 9th, 1:00 PM Apr 9th, 3:00 PM

A Comprehensive Survey of Agentic AI: Design Principles, Security Risks, and Ethical Consideration

Face-to-face

In the past several years, the world has managed to transition away from simple automation to independent AI systems. Agentic AI is an agent that can work independently, carrying out all essential plans and implementations without any kind of supervision from a human being. This review has tried to demonstrate the transformative impact that Agentic AI brings to contemporary models of intelligence by means of synthesis of perception, reasoning, and goal. We utilized the phrases Agentic AI, autonomous AI, multi agent systems as keywords in Google Scholar, ScienceDirect, arXiv, and other digital libraries. We have used these 38 main papers published between 2023 and present as part of search process. This paper has taken a deep, close approach to Agentic AI systems, thereby discussing their technical foundations, challenges, applications, and societal impacts. In this paper, we discuss the different aspects of Agentic AI, including the technical side, challenges, applications, and impacts. In this study we found Agentic AI systems are capable of a 34.2 % reduction in the time of task execution, a 7.7 % increase in accuracy, and a 13.6 % improvement in resource utilization. In this survey, we investigate the performance and benchmarking of model architectures, security, privacy, and human-in-the-loop integration; and analyze frameworks for efficient human-AI collaboration. This survey draws on ideas from recent publications, equipping researchers with a holistic view of the state, challenges, and directions for future research necessary to keep Agentic AI safe, aligned, and beneficial to society.