Forrester：企业 AI 治理面临战略脱节问题

The good news: AI governance is on the enterprise agenda. The problem: It still isn’t embedded deeply enough in enterprise AI strategy.

According to Forrester’s most recent data, only 42% of AI decision-makers say governance strategy is included in their organization’s official AI and agentic AI strategy. Policies may be proliferating, but strategic integration is still catching up. That gap matters as enterprise AI moves from search and content creation into agentic systems capable of taking actions at scale.

For CTOs and CAIOs, the challenge is moving from governing AI use to building governance into how the enterprise buys, builds, deploys, and scales AI. Effective, appropriate governance is critical to ensuring that AI remains an asset and not a liability.

Your Organization Has AI Policies, But It Needs A Governance Strategy

Documenting AI policies is a good place to start but hardly the finish line. Policies can define acceptable use, clarify responsibilities, and establish protections. Having individual policies doesn’t mean that an organization has a coherent governance strategy, however. The need for one becomes obvious when enterprises scale AI. AI decision-makers need repeatable ways to decide which use cases warrant investment, the risks that they will accept, what suppliers must demonstrate, who can approve deployments, and where human oversight is mandatory.

Agentic AI Raises The Governance Stakes

Governing human use of AI is hard enough, but the arrival of autonomous AI agents only raises the complexity and the stakes. An AI agent that accesses enterprise data, executes business processes, or interacts with external applications creates a different governance challenge than a standalone generative AI tool. With the expanding use of agentic AI, organizations must govern not only what a system produces but also what it can do, what it can access, when humans should intervene, and who is accountable.

Having A Governance Strategy Is Not The Same As Governing AI

A governance strategy defines what an organization intends to govern, why, and according to which principles, policies, and decision rights. Governing AI turns that strategy into action through the decisions, processes, and controls that shape how AI is built, bought, deployed, and used.

AI governance that exists only in policies and review committees won’t keep pace with AI-assisted work. Organizations must translate governance strategy into the processes and decisions through which AI is developed and used, embedding governance across development, procurement, deployment, and oversight of AI and agentic workflows. Governance cannot be a compliance check bolted onto the end of an existing process. Questions about data access, security, trust, accountability, and business value must shape AI investments from the outset.

Good Governance Focuses On Outcomes

Enterprises don’t need to choose between moving quickly with AI and governing it effectively. But they do need to stop treating governance primarily as a defensive mechanism for constraining AI. The goal isn’t more governance but governance that helps enterprises make better decisions at scale.

• Design governance for action, not just output. Define authority, access, oversight, and escalation before agents enter critical workflows. Extend the same requirements to AI acquired from suppliers.
• Put governance into the AI lifecycle. Apply governance when teams select, build, procure, deploy, and monitor AI, not as a review added just before production.
• Treat AI literacy as a governance capability. Give employees practical, role-specific guidance for using AI safely and effectively, from responsible development and evaluation for technical teams to knowing when to question AI outputs and apply human judgment for nontechnical employees.

Forrester clients can explore the full trends to understand how enterprises are strengthening AI governance and connecting it with AI outcomes and workforce readiness: The State Of AI Governance, 2026: From Policy Adoption To Strategic Integration.

We’re always interested in hearing your perspective on AI governance. Please feel free to schedule a guidance session or inquiry call.