에이전트형 AI : 최신 동향과 기회

Agentic AI Transforming the Enterprise Value Curve

Agentic AI is redefining the enterprise AI landscape by moving beyond traditional and generative models toward autonomous systems that can reason, plan, and act across complex workflows. Enterprises are increasingly moving from experimentation to production-grade AI deployments, embedding agents within workflows, customer journeys, and IT operations. This report explores key trends such as task-specific AI agents, multi-agent collaboration, enterprise integration approaches, and trust & safety frameworks, driving real-world adoption.

Vendors like UiPath, Zoho, Microsoft, and ServiceNow are enabling agent deployments across ITSM, customer support, finance, and HR. Enterprises are increasingly integrating these agents into production environments, leveraging APIs, orchestration layers, and hybrid strategies for scalability and control.

As the agent economy matures, stakeholders who invest early in lifecycle orchestration, trust frameworks, and embedded integration will gain a competitive advantage. This report offers strategic insights and real-world use cases to help business leaders lead in the Agentic AI era.

The Impact of the Top 3 Strategic Imperatives on the Agentic AI Industry

Disruptive Technologies

Why

• Agentic AI is fundamentally disrupting how AI systems operate by enabling them to autonomously perform tasks, interact and make decisions, without constant human oversight.
• This is a significant leap from traditional AI systems that primarily focus on content generation. Agentic AI is particularly appealing to businesses because it significantly reduces labor costs and enhances productivity by automating complex tasks.

Frost Perspective

• Platform vendors can develop tools and frameworks that enable agents to interact with enterprise systems and external services through APIs, pre-built accelerators and integration layers. Also, they can offer agent orchestration platforms where multiple specialized agents collaborate on complex tasks.
• Opportunities exist for service providers to develop specialized infrastructure services and bespoke agents, integrate agents into enterprise workflows, and build governance and security services to ensure compliance.

Geopolitical Chaos

Why

• Ongoing friction between global economies has led governments to introduce sanctions and trade tariffs and reduce dependence on technology (hardware and software) developed overseas.
• This has led governments worldwide to push for homegrown computing infrastructure and AI development with strict localization mandates.

Frost Perspective

• Sovereign AI is a result of rising digital protectionism, a shift where AI infrastructure, data, and talent are seen as national security assets rather than just technological capabilities.
• Opportunities span the AI ecosystem, from creating localized datasets and local compute infrastructure, to building region-specific AI models, to offering services to help enterprises adapt global governance and compliance.

Internal Challenges

Why

• Enterprise AI implementation continues to be hindered by data fragmentation, with many enterprises struggling to establish a unified data foundation.

Without a unified, high-quality, and real-time data infrastructure, AI models lack the comprehensive datasets needed to generate accurate and actionable insights.

Frost Perspective

• As AI evolves, managing the data lifecycle has become as critical as the AI models.

Opportunities for technology vendors and service providers exist in offering data services, such as labeling tasks and synthetic data generation, data management (i.e., integration of diverse data sources into unified pipelines), and data monitoring services to ensure data health.

Growth Drivers

Efficiency improvements and cost reductions represent compelling economic drivers for agentic AI adoption

• Ability to extract actionable value from the growing volumes of enterprise and customer data drives Agentic AI uptake
• Increasing availability of enabling technologies and infrastructure

Growth Restraints

• Lack of trust slows enterprise adoption
• Clear return on investment (ROI)
• Lack of leadership commitment
• Lack of clarity concerning regulatory frameworks and ethical practices

Table of Contents

Agenda

• Strategic Imperatives
• Why is it Increasingly Difficult to Grow?
• The Strategic Imperative 8™
• The Impact of the Top 3 Strategic Imperatives on the Agentic AI Industry

Growth Opportunity Analysis

• Glossary
• With AI, the Only Constant Is 'Change'
• AI: A Technology Priority for Global Enterprises
• Evolution of AI Systems: Traditional vs Generative vs Agentic
• What is Agentic AI?
• Key Characteristics of Agentic AI
• Growth Drivers
• Growth Restraints
• Agentic AI Tech Stack
• Agentic AI Tech Stack-Explained
• Emerging Agentic AI Trends
• Task-Specific AI Agents: An Introduction
• Task-Specific AI Agents: Emerging Use Cases
• Task-Specific AI Agents: Agentic AI Deployments Across Key Industry Sectors
• Task-Specific AI Agents: Agentic AI Deployments Across Other Sectors
• Key Examples of Agentic AI Deployments
• Emergence of Multi-AI Agent Collaboration Systems
• Multi-AI Agents: Understanding Different Approaches
• Emergence of Multi-AI Agent Collaboration Systems: Key Vendor Ecosystem
• Enterprise Integration of Agentic AI
• Trust and Safety: Data Concerns and Ability to Assess ROI Continue to Challenge AI Adoption
• Trust and Safety: Agentic AI Introduces a New Category of Risks, Going Beyond Traditional IT Security Threats
• Trust and Safety: Recommended Approaches for Threat Mitigation
• Emerging Business Model: Outcome-Based Agent-as-a-Service

Companies to Action

• Key Companies to Action: Microsoft
• Key Companies to Action: ServiceNow
• Key Companies to Action: Zoho
• Key Companies to Action: UiPath

Growth Opportunity Universe

• Growth Opportunity 1: Agentic AI Services

Appendix & Next Steps

• Benefits and Impacts of Growth Opportunities
• Next Steps
• List of Exhibits
• Legal Disclaimer