51% of Global Capability Centers Stuck in AI Pilot Stage — New Playbook Maps the Path to Scale

A new joint report from Zinnov, nasscom AI, and Tiger Analytics finds that 51% of Global Capability Centers (GCCs) remain stuck at the earliest stages of AI maturity — running AI in isolated pockets without enterprise-wide ownership or measurable impact.

The report, based on interviews with 75+ GCC leaders, identifies a repeating failure pattern: AI charters anchored around technology rather than enterprise priorities, budgets fragmented across business units, operating models chosen by default, and governance that arrives after programs have already drifted off course.

Why AI Programs Stall in GCCs

Scaling AI in a GCC is not a single challenge — it is a sequence of challenges where order matters:

• Charter must be right before capability investment pays off
• Capability must be built before scale can hold
• Governance must run alongside execution — not after it — covering AI risk, adoption rates, and LLM token costs as programs expand

When organizations skip or rush these steps, even well-funded initiatives stall.

What the Playbook Covers

• Aligning the AI agenda to enterprise priorities (not technology)
• Structuring budgets to fund both delivery and exploration
• Deciding when to build internally vs. partnering externally (and ensuring partnerships transfer capability rather than create dependency)
• Building domain expertise into delivery teams alongside engineering depth

The report's central question for GCC leaders: Is your GCC structured to move from execution to ownership, or still waiting for the mandate to arrive?

Read the full report on Zinnov