에이전트형 기업의 보안 : 제2부 - 주요 보안 도메인에서 에이전트의 영향 평가

This IDC Perspective examines how agentic AI transforms enterprise security, requiring a shift from static controls to continuous, identity-driven governance. It outlines new risk categories - external, internal, and toolset-based - and emphasizes the convergence of security domains as agents operate autonomously. The updated IDC framework advocates for adaptive policies, real-time monitoring, and integrated controls to maintain resilience and compliance. Actionable recommendations guide technology buyers in strengthening identity-centric controls, deploying full-fidelity telemetry, and designing for ongoing disruption."Agentic AI compels enterprises to rethink security: while core principles endure, the real challenge is self-inflicted disruption from poorly governed agents. Continuous, identity-driven governance and operational resilience are now essential for navigating a landscape where disruption is constant, and boundaries are fluid," says Duncan Brown, group vice president, IDC.

Executive snapshot

• Key takeaways
• Recommended actions

Situation overview

• An update to IDC's Cybersecurity Capabilities Assessment Framework
• Domains of IDC's Cybersecurity Capabilities Assessment Framework 2.0
  • Network security
    • Network security changes by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in security toolsets
  • Endpoint security
    • Endpoint security changes by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in endpoint security toolsets
  • Identity and digital trust
    • Identity and digital trust change by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in identity and digital trust toolsets
  • Data security
    • Data security changes by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in data security toolsets
  • Application security
    • Application security changes by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in application security toolsets
  • Response, recovery, and resilience
    • Response, recovery, and resilience changes by risk class
      • External agentic threats
      • Internal agentic threats
  • GRC
    • GRC changes by risk class
      • External agentic threats
      • Internal agentic threats
      • Agents in GRC toolsets and services
      • GRC in the context of AI Governance

Advice for the technology buyer

Learn more

• Related research
• Synopsis