The AI talent paradox: A catalyst or crisis for GCCs?

GCC talent market today. The combination of technical expertise fused with domain knowledge is what hiring managers across the GCC sector are finding it difficult to hunt.

According to fresh estimates by staffing firm Xpheno, out of 28,000 to 30,000 Indian tech professionals with genuine enterprise AI exposure, only 4,000 to 5,000 qualify as true AI specialists, who have hands-on engineering skills, solutioning, and architecture depth.

“The gap is real and it’s nuanced. It’s not that AI talent doesn’t exist in India, it does, and in significant volume. The challenge is finding professionals who combine strong technical foundations in machine learning and data engineering with the ability to apply those skills to specific domain problems; in our case, payments, billing, and regulated financial workflows,” Biju Davis, senior vice-president and India site leader at InvoiceCloud told ETGCCWorld.

Where does the gap lie?

Staffing firm TeamLease Edtech puts the GCC talent gap in AI and data analytics roles at 38-42 per cent, and notes that for every ten open AI positions, only about six suitable candidates are available.

Notably, recruitment firm Grafton reports that AI-focused searches are now taking 30-50 per cent longer to close than traditional technology roles.

But the harder problem is not one that traditional bulk hiring alone can solve.

“What we encounter most often are candidates who are strong in theory but have limited experience taking AI from proof-of-concept to production-grade systems. Building AI that works in a demo is a very different problem from building AI that operates reliably at scale, handles edge cases gracefully, and meets the auditability requirements of regulated industries. That specific combination–technical depth plus production experience plus domain context–is where the shortage is most acute, and it’s what takes the longest to hire for,” Davis explained.

When asked to rank the gaps, Davis places practical experience first as the hardest challenge, ahead of domain knowledge and even technical skills.

While the reasoning is counterintuitive, he noted how it’s shared across the sector.

“Technical skills are actually the most teachable. A strong engineer with solid fundamentals can learn a new framework or model architecture relatively quickly. What’s harder to develop fast is the judgement that comes from having shipped AI systems in production, i.e knowing when a model is good enough, when to trust it and when not to, how to instrument it for observability, and how to explain its behaviour to non-technical stakeholders,” Davis said.

Manish Goyal, managing director and head of GCC at Alvarez & Marsal, views the challenge through an organisational lens. The talent and skills are available, he argues, but they remain fragmented across functions and teams.

“The biggest gap today is not in foundational AI skills alone, but in the ability to combine technical expertise with industry knowledge and real-world implementation experience. The required skills exist in silos across different teams. The real gap is the execution excellence that can effectively connect the technical-functional-consulting expertise and translate that into a practical solution. On top of this you need effective change management and value-tracking mechanism to achieve measurable outcomes,” Goyal told ETGCCWorld.

Kamal Karanth, co-founder of specialist staffing firm Xpheno, offers an explanation for why that execution gap persists. The best and most complex AI systems, he pointed out, were built outside India. However, that history shows up in the talent pool today.

“The top 10 companies by volume of active AI engineering specialists hiring worldwide all have a presence in India through affiliates or GCC units, yet only 5 per cent of their active AI engineering demand is directed at India. This alone points to a confirmed competitiveness gap–not just in talent supply, but in the overall readiness of the talent pool,” Karanth noted.

Furthermore, he adds a timeline dimension that is often easy to overlook. Even a professional with 15-20 years of total experience will bring, at most, six years of genuinely relevant AI exposure.

Rethinking in hiring

The pressure on talent supply is changing the logic of hiring. Degrees from premier institutions still carry weight, but they are no longer the primary filter. What matters more now is evidence of systems built, problems solved, and models deployed under real conditions.

At InvoiceCloud, the interview process has also been rebuilt around that principle.

The focus is on evidence of applied work–open-source contributions, projects that shipped, systems that ran under real load. He notes how the GCC scouts for a professional capable of fine-tuning a model for a specific business problem, built a RAG pipeline, or worked with agentic AI workflows in a production context.

“Certifications are a signal of intent and structured learning, which matters. But they’re a starting point in our evaluation, not an endpoint. The interview process we’ve built is deliberately oriented around real problems; we want to see how people think and build, not just what they’ve studied,” Davis said.

Maya Nair, executive director at Grafton Recruitment India, sees the same shift across GCC clients her firm works with, and points to one important marker of how far the pendulum has swung.

“Five years ago, academic pedigree often played a significant role in screening candidates. Today, most hiring managers are far more interested in what a candidate has built, implemented, or solved than where they studied. Their public portfolios gain interest and many hiring managers want to see their contribution to solving a problem in a project,” Nair explained.

From a recruitment perspective, the AI skills gap is one of the most visible challenges facing GCCs today.

“Over the last 18-24 months, we have seen a significant increase in demand for AI, generative AI, machine learning, data science and AI product roles, while the available talent pool has grown at a much slower pace,” she added.

According to Karanth, the proportion of entry-level hires in GCCs has nearly doubled over the past 12 months, compared with a historical preference for experienced lateral talent.

“For years, GCCs relied heavily on experienced professionals to hit the ground running. GCCs have not been big fans of the talent BUILD model that involves onboarding and training freshers and entry-level talent. However this is seen changing now with GCCs opening up their hiring plans to absorb this talent. The entry-level talent cohort among Indian GCCs–those with under two years of experience–has nearly doubled over the last twelve months,” Karanth said.

That build-versus-buy calculation is a reshaping strategy at the institutional level too. At Accenture, Paul Jeruchimowitz, senior managing director and global GCC practice lead, argues that recruiting alone cannot close the gap, and that the data from leading centres confirms it.

“Our research shows 74 per cent of GCCs cite talent gaps as their number one barrier to innovation. However, a recruiting strategy alone won’t solve this. Increasingly, the focus is shifting from talent acquisition to talent transformation–redesigning work for AI, creating new roles and building learning programs tailored to the GCC’s specific context and talent,” Jeruchimowitz said.

Sandeep Kohli, deputy global vice-chair for talent, EY GDS describes a workforce strategy rebuilt from static job definitions toward dynamic, capability-led roles, with AI doing some of the building work itself.

“We are leveraging the power of AI and GenAI to reshape how we identify, develop, and grow talent, using data-driven insights to better understand skills, potential, and aspirations. Through predictive analytics, we are able to anticipate future skill needs and proactively invest in closing those gaps. Importantly, our approach is rooted in inclusion and scale, with 100 per cent of our people enabled with AI training,” Kohli told ETGCCWorld.

New York-based banking major BNY views the challenge as making AI accessible across the organisation. Suswar Ganu, head of engineering and global co-head of sales, marketing & distribution technology and site executive for India for BNY, said the focus is on building AI fluency across the workforce.

“While demand for specialized AI talent continues to grow, we are equally focused on upskilling and developing internal capabilities by giving employees broad access to AI tools, training and practical learning opportunities that help them apply AI in their day-to-day work. Our commitment is to democratise AI responsibly, enabling employees of all technical backgrounds to build fluency and work more effectively alongside AI,” Ganu said.

The mismatch, however, begins before candidates ever walk into an interview. Arindam Mukherjee, co-founder and CEO of NextLeap, which works closely with AI-led GCC hiring pipelines, traces it back to what is being taught versus what is being demanded.

“There exists a significant mismatch between what learners learn and what the market expects of them. While the popularity of AI courses is steadily increasing, there is also a growing need among employers for AI practitioners who can analyse data sets, develop practical AI-based solutions, consider business context and team collaboration issues, and more. The current job market calls for practitioners rather than learners–people who have both the necessary technical background and experience in solving actual business problems,” Mukherjee said.

Focus on skills-first roles

The widening gap between the demand and supply of AI talent is pushing organisations from role-based hiring to a skills-first and outcome-led approach.

“The real differentiator is the ability to turn meaningful insights into actionable outcomes that deliver business impact and ultimately better customer experiences. Academic qualifications apart, there is a strong emphasis on skills like problem framing, product thinking, hands-on experience for better judgment, learning agility, and the ability to integrate rapidly evolving AI systems,” said Amit Vaish, VP and head of people team, Optum India.

Shantanu Rooj, founder and CEO of TeamLease Edtech, points to flexible hiring as evidence that the market has already absorbed this reality.

He revealed how nearly a quarter of niche AI roles are now being filled through contractual or contract-to-hire routes.

“GCC hiring is clearly becoming more skills-led. Degrees still matter as a foundation, especially for advanced technical roles, but they are no longer sufficient proof of job readiness. GCCs are increasingly prioritising portfolios, certifications, project work, problem-solving ability and hands-on exposure over pedigree alone. This is a strong signal that employers are buying capability, not just credentials,” Rooj said.

The geographic question, on whether Tier II cities can realistically expand the pool, surfaces in most conversations. However, industry experts differ in their assessment of its near-term potential.

Teamlease’s data shows Tier II cities currently accounting for just 10-12 per cent of GCC recruitment, with complex AI mandates continuing to return to metros because depth is not yet available at the periphery.

“To fill this gap, cities like Coimbatore, Ahmedabad., Indore, Kochi, Bhubaneswar, Jaipur, Vishakapatanam and Mysuru are producing strong technology talent and benefiting from improving digital infrastructure, quality educational institutions, and a growing startup ecosystem,” Nair noted.

Interestingly, Davis offers a note of caution that cuts across the geography debate.

“What I’d caution against is treating Tier II hiring as a cost strategy. The best reason to look beyond traditional hubs is access to talent that’s hungry, capable, and less competed-for–not to find cheaper alternatives. At InvoiceCloud, our mandate is to build a world-class centre in Hyderabad, and that means hiring on capability wherever it exists,” Davis said.

Goyal at A&M draws the longer map that capability building is now as strategically significant as talent acquisition, perhaps more so.

“The pace at which AI technologies are evolving requires organisations to continuously build and refresh capabilities. We see the future as a combination of targeted hiring, continuous learning, and hands-on experience through global client engagements, enabling teams to build expertise while delivering measurable outcomes,” Goyal added.

Jeruchimowitz, ending the argument plainly, said, “The most valuable talent will be those who can combine AI fluency with deep industry and functional expertise, and the advantage will not come from technical capability alone.”

For GCCs aiming to sustain their advantage, the talent paradox may ultimately prove to be a turning point. While the gap may be widely viewed as a crisis, for GCCs willing to invest in capability building, it could become a catalyst for a more sustainable talent model.