AI in the workplace.

Most of what's written about AI and work is forecast. We write from what's actually showing up in the roles we fill.

How AI is changing the shape of GCC work, the skills that suddenly matter, and the senior roles being created and quietly retired. Read from live mandate data across India's GCCs, not from a trend deck.

Updated May 2026 Written by Sachith Rai, Recruise editorial

Key takeaways

AI is changing job composition before it changes headcount. The tasks inside a role shift first, which is why org charts look stable while the skill bar quietly rises.
The scarce hire is no longer the person who can build a model, it's the leader who can decide where AI belongs in the workflow and where it doesn't.
GCCs that treat AI as a tooling rollout under-hire. The ones that treat it as an operating-model change re-shape their senior bench.

The pattern

AI changes the tasks first, the titles last.

Recruise mandate data · 2025–2026

The role looks the same on the org chart. The work inside it doesn't.

When a function adopts AI seriously, the first thing that moves isn't the headcount, it's the mix of work inside each role. The routine production shrinks, and the judgment, review, and orchestration grow. A title that meant one thing two years ago now asks for a different person.

That's why hiring plans built off last year's job description miss. The req still says the old role. The work needs someone who can lead a team where AI does the first draft and people own the call. Reading that shift early is the difference between hiring for the role you had and the role you now have.

The scarce skill is judgment about where AI belongs.

The market is full of people who can use the tools. What's rare is the leader who can decide where AI genuinely improves an outcome and where it quietly degrades it, and who can hold that line against the pressure to automate everything because it's possible.

For a GCC, that judgment lives in the senior bench. It's the head of a function deciding which decisions stay human, which get assisted, and which get handed over. Hire for the tooling and you get speed. Hire for the judgment and you get speed that doesn't cost you the quality the parent is watching.

"The question isn't whether AI can do the task. It's whether the outcome survives handing it over. The leaders who can tell the difference are the hire that matters now."

Sachith Rai · MD & Founder, Recruise

How we know what we know

This read is drawn from the roles we're actually filling, not a trend forecast.

60+ GCCs

Active mandate relationships across IT, BFSI, Pharma, and ER&D, where we watch role composition change in real time.

4,000+

Placements since 2006 behind our read on which skills are rising and which are quietly being retired.

Live

We write from mandates in flight, so the read reflects what's happening this quarter, not last year's survey.

One pattern worth knowing, every week.

The Signal is our weekly read on the senior GCC talent market, including how AI is reshaping it. One chart, one pattern, no noise.

Subscribe to The Signal

Frequently asked

Questions on AI and hiring.

How AI is changing which roles a GCC needs, and how to brief for them.

What does AI change about hiring for a GCC?

AI shifts the senior bench from people who run process to people who design how the work gets done. The roles that grow own judgment, architecture, and review. The ones that shrink were built around routine execution. Hiring well now means reading that shift before the org chart does.

Are GenAI hires the same as traditional AI and ML hires?

No. A traditional AI or ML hire builds and trains models. A GenAI hire applies foundation models to a product or workflow, which is a different skill set and a different talent market. Most job descriptions blur the two, and the wrong brief draws the wrong shortlist.

Why do most GCC AI pilots stall at the same point?

They stall when a tooling rollout meets an operating model that was never redesigned for it. The pilot proves the technology works, then stops because no one owns the change. Centres that treat AI as an operating-model question, and hire for it, get past that wall.