The New AI Operating Model: What Indian CEOs Must Get Right

News

1. AI Skills May Shape Appraisals Over Next Three Years

2. Anthropic Tests Marketplace Where AI Agents Negotiate Deals

3. AWS Reaffirms $12.7 Billion India Cloud Push

4. DeepSeek Previews V4 Model for Agentic AI Tasks

5. Reliance Meta AI Venture Names Parminder Singh CEO

6. Meta Layoffs, Microsoft Buyouts Signal AI Reset

India’s largest enterprises have spent three years running artificial intelligence pilots. The results have been promising enough to keep boards interested and investors engaged, but often too limited to justify the infrastructure costs behind them or deliver the enterprise-wide transformation chief executives promised.

As companies push AI deeper into operations, from credit decisions and hospital triage to factory floors and retail supply chains, a harder question is now confronting boards and management teams: why are so few pilots scaling?

The answer, according to chief executives across India’s technology, education, staffing and infrastructure sectors, is that the problem was never primarily technological, but structural. The challenge is not adoption but operationalization, and closing that gap requires rethinking how companies are built, led and governed, not the software stack they run.

PwC’s 29th Global CEO Survey, conducted in late 2025 and drawing on responses from 4,454 chief executives across 95 countries, quantifies the stakes. Companies reporting both cost and revenue gains from AI were two to three times more likely to have embedded the technology extensively across products, demand generation, and strategic decision-making, not just in isolated pockets. The findings suggest that AI has become a defining fault line for growth and profitability. Three imperatives — platform architecture, workforce readiness, and governance design- separate organizations extracting durable value from those still cycling through pilots.

“Winning in the age of AI requires a fundamental shift in thinking,” said Pankaj Vyas, Managing Director and CEO of Siemens Technology and Services. “It’s no longer about isolated pilots but acting with urgency to redesign the entire enterprise around intelligence.”

Operating Stack

The architectural shift most organizations cannot make, several CEOs argue, is fundamental. Shridhar Mantha, CEO of Generative AI Business Services at Happiest Minds, describes it as the move from project thinking to platform thinking, from deploying AI to solve a defined problem to building a unified stack in which data, models, orchestration layers and governance frameworks are tightly integrated and continuously evolving.

Breaking down silos, both technical and organizational, is a precondition, not an afterthought.

“The most effective strategy is to anchor AI initiatives to measurable business outcomes,” Mantha said, “ensuring that every use case deployment feeds into a larger, continuously evolving enterprise capability rather than remaining an isolated success.”

Milind Shah, Managing Director of Randstad Digital, frames the same inflection point in terms of enterprise architecture. The shift to an AI-first economy, he argues, demands a wholesale redesign of how enterprises compete and create value.

The organizations best positioned to capitalize on it are those that can orchestrate technology, talent and governance as a single unified system.

“Competitive advantage will not come from access to technology,” Shah said, “but from the ability to operationalize it with speed, discipline, and strategic coherence across the enterprise.”

Max Liu, Co-Founder and CEO of TiDB, argues that the CEO’s role itself is being redefined in the process.

In an AI-first economy, chief executives are no longer primarily strategy-setters and execution overseers but system designers, responsible for ensuring that AI is embedded coherently across workflows, customer experiences and decision-making processes.

“CEOs must act as stewards of trust,” Liu said, “ensuring responsible AI use, ethical guardrails, and transparency in decision-making.”

Readiness Gap

Beneath the infrastructure challenge lies a deeper one, and it is the one that most organizations are least equipped to address. Nearly two-thirds of stakeholders describe the pace of AI-driven change as very fast or extremely fast, according to Pearson research published in early 2026.

Yet only a quarter believe their organizations are keeping pace, and more than half of employers report difficulty finding workers with the skills to deploy AI effectively.

Vinay Kumar Swamy, Country Head of Pearson India, argues that the problem is being systematically misdiagnosed. Organizations treat AI readiness as a matter of tool access or basic literacy, only to discover that access alone yields nothing.

True readiness, he contends, is the ability to apply AI with judgment, responsibility, and purpose, combining functional proficiency with ethical stewardship and the kind of contextual intelligence that machines cannot replicate. 

“The constraint is not access to technology,” Kumar Swamy said. “It is readiness.”

Learning and AI deployment must happen simultaneously, embedded directly into workflows rather than delivered as a training program after the fact. As skills half-lives shrink to two or three years, readiness can no longer be defined at the point of hiring. It must be continuously built, measured, and refreshed in the flow of work.

Six in ten employers already say they prioritize a balance of human and technical skills when hiring, ranking communication, adaptability and collaboration alongside AI proficiency, according to the same Pearson research. Yet organizations continue to over-invest in tools while under-investing in the human capabilities required to guide and evaluate what those tools produce. “Productivity gains emerge at the point of partnership, not substitution,” Kumar Swamy said. The leadership implication is structural, not merely managerial.

Chief information officers, chief technology officers, and chief human resources officers can no longer operate as separate functions with loosely aligned mandates. The AI era demands that they function as a single system, translating technology adoption into workforce performance and organizational value. Without that alignment, AI investment stalls, not because the technology fails, but because the organization around it does.

Trust by Design

As AI systems become more agentic, capable not just of generating recommendations but of initiating actions and operating across extended time horizons without human intervention, the governance question moves from the legal department to the design table.

Who is accountable when an autonomous system makes a consequential error? How is performance defined when the system is continuously learning? These are no longer hypothetical questions for India’s largest enterprises. They are operational ones.

Narendra Sen, Founder and CEO of RackBank and NeevCloud, argues that the organizations best equipped to answer them are those building what he calls hybrid super-teams, or structures in which domain experts and AI systems co-evolve, supported by governance that treats ethics and risk not as constraints on deployment but as the foundation on which institutional trust is built.

“The real competitive edge doesn’t just come from the technology itself,” Sen said, “but from the velocity of our adaptation.”

Vyas of Siemens echoes that framing, stressing that as AI evolves toward more agentic systems capable of reasoning and acting autonomously, the ability to scale becomes paramount, and scaling demands not just speed but deep collaboration, anchored by ecosystems that can translate experimentation into enterprise-wide impact. 

“Embed AI across the business, not just in pockets,” Vyas said. “We must align talent and data strategies to our core objectives and build systems that keep human expertise at the center.”

Together, those three imperatives, platform architecture, workforce readiness and governance design, separate organizations extracting durable value from AI from those still cycling through pilots.

According to Nasscom’s AI Adoption Index, published in 2025, 87% of Indian enterprises now use AI in some form, with banking, healthcare, retail and manufacturing accounting for roughly 60% of the value created. The adoption numbers are no longer the story. What happens after adoption is.

Editor’s Note: MIT Sloan Management Review’s AI Research Forum will make its India debut in Bengaluru on 23 July, bringing together enterprise leaders, researchers, and practitioners to examine how autonomous AI is moving from experimentation to governed deployment at scale. To speak, partner or attend register here