How AI Is Changing What a Global Capability Center Does
AI in global capability centers is usually discussed as a jobs story: how many roles go and how many stay. The more useful story is what the work itself is becoming. That is where the data, and the decisions leaders face, are actually moving.
A global capability center, or GCC, is a unit a multinational owns and runs offshore to do work for the parent company. For the earlier shift from savings to ownership, see why India's GCCs stopped competing on price.
From Back Office to AI and Engineering Work
The mandate changed before the headlines did.
A Zinnov report covered by Business Today found that 96% of India's GCCs set up since FY21 launched with a product, R&D or engineering mandate. Those centers also reach maturity in under five years, against five to ten for older ones.
The same report found that generative and agentic AI now account for 52% of AI use cases in these centers. It is the first time they have overtaken traditional AI and machine learning.
EY's GCC Pulse Survey of India GCC leaders points the same way. In it, 92% of leaders say their centers now deliver more than cost savings, and 58% are investing in agentic AI, with another 29% planning to start within a year.
What AI in Global Capability Centers Means for Jobs
Estimates of job loss vary widely, which is the first thing to notice.
Business Today's reporting shows how far apart the numbers are. One analyst puts this year's cuts at 25,000 to 30,000 roles, while another estimates 4,000 to 5,000. The article also ties part of the cuts to parent companies restructuring, not only to AI.
Looking further out, a report from the Observer Research Foundation models gross AI-related displacement of about 40,000 to 150,000 GCC jobs through 2030, depending on how fast adoption moves. It says new AI-augmented roles reduce the net loss but do not eliminate it, and it sees fewer entry-level, high-skill openings as the bigger risk than sector collapse.
The pattern is a shift in mix. Routine work shrinks, while engineering, product and AI governance work grows.
Why Scaling AI Is the Hard Part
Running AI pilots is easy. Scaling them is where centers struggle.
EY's survey shows the foundations lagging ambition. While 58% of centers invest in agentic AI, 67% are still formalizing their data strategies, and only 7% have a fully embedded cybersecurity center of excellence. The ORF report likewise says the gains depend on closing gaps in skills, data governance and infrastructure.
None of those is a model problem. They are operating model problems, which is why the centers that scale AI tend to look different from the ones that only run pilots.
What Leaders Should Watch
Four questions separate a center that uses AI from one that runs on it.
Ask where the data lives and who owns it. Ask who is accountable for AI governance and security, since EY found only 7% of centers with a fully embedded cybersecurity center of excellence. Ask which work is routine and which is complex, and plan the skill mix around that split.
Finally, ask how success is measured. A center measured on outcomes behaves differently from one measured on headcount, a point we explored in our conversation on GCCs beyond cost.
What We Have Seen
The causes of stalled AI work show up at setup, not at launch.
NSquare offers GCC setup consulting, covering strategy, location selection, legal and operational setup, with a typical engagement of three to six months. We also run GCC operations, including talent acquisition and delivery, from a Frisco and Pune onshore-offshore model with SLA-backed delivery.
The gaps EY and ORF describe, such as data strategy and governance, are the same ones we raise at the start of a setup. Our Pune offshore center is also where we recorded a CXOTalks conversation with Adani Enterprises on where GCCs go next.
FAQs
Short answers to the questions leaders ask first.
What is a global capability center?
A unit that a multinational owns and operates, often in India, to run technology, engineering and business work for the parent company.
How is AI changing GCCs in India?
Newer centers launch with product and engineering mandates, and generative and agentic AI now make up more than half of AI use cases. Routine work is shrinking while higher-value work grows.
Will AI eliminate GCC jobs?
Estimates vary widely, from a few thousand roles this year to scenarios of up to 150,000 through 2030. Reports tie the cuts to both AI and parent-company restructuring, and new roles reduce the net loss without erasing it.
What holds GCCs back from scaling AI?
EY's survey and ORF's report point to data strategy, governance, security and skills gaps, not to the AI models themselves.
Written by the NSquare Xperts team.
