When a CEO announces that AI could cut project costs by a quarter, most headlines treat it as a technology story. It is not. It is a confession.
R. Srikrishna, CEO of Hexaware Technologies, recently told the press that AI integration could reduce IT project costs by as much as 25%. The statement arrived without granular detail, without methodology, without a timeline. Just a number, floating in the ether of earnings season optimism. The market received it as forward-looking guidance. I received it as something far more revealing—a public admission that the business model underpinning a trillion-dollar industry has quietly begun to dissolve.
Follow the money, not the noise. What Srikrishna actually announced was not an efficiency gain. It was the beginning of a transfer of wealth from labor to capital, from the thousands of engineers logging hours to the shareholders who will watch the cost-per-delivery curve bend downward.
The global IT services market sits at roughly $1.2 trillion annually. India's top firms—TCS, Infosys, Wipro, HCL—employ close to five million people, making this the largest concentration of white-collar technical labor in the world. For three decades, the economics were simple: hire engineers at a fraction of Western salaries, bill clients per-hour, and maintain gross margins of 30% or more. The model was arbitrage. Pure and simple.
AI breaks the arbitrage. When coding, testing, documentation, and even project management can be partially automated, the per-hour cost basis collapses. Human capital becomes optional at the margins. And when that happens, the billing structure—the sacred "time and material" framework—loses its meaning. A 25% cost reduction doesn't come from optimization. It comes from substitution.
What Srikrishna failed to mention is that if costs drop by 25% while revenue declines by the same amount, profit margins do not improve. They stay flat at best, and deteriorate at worst. The CEO's mention of "balancing productivity gains with pricing strategies to maintain margins" was not a nuance. It was a confession.
The Defensive Posture
Let me be precise about what is happening here. The 25% figure isn't a technical measure. It is a preemptive signal to the market. Every enterprise client negotiating a new contract now knows that the vendor's costs can be reduced by 25% with AI. That knowledge shifts the negotiating table. Clients will demand a share of the reduction, either as lower rates or as value-added services at no additional cost. If Hexaware doesn't voluntarily acknowledge the reduction, the client will either go to a competitor or procure AI tools directly.
This is the strategic reality behind the announcement. The vendor is not the aggressor; it is the defender, hoping to maintain client relationships by acknowledging the inevitable.
Based on my experience auditing IT service contracts in cross-border markets, I've observed a consistent pattern: every time a cost-cutting innovation emerges, the vendor who announces it first captures short-term goodwill but forfeits long-term pricing power. The clients remember the number. They do not forget it in negotiations.
Industry-Wide Unraveling
The full scope of this narrative extends far beyond a single company. The 25% reduction, if realized across the industry, translates to roughly $240 to $300 billion in annual value creation. That value does not accrue to the IT services firms. It splits between clients (through price reductions), AI tool providers (through licensing fees), and, eventually, to a smaller pool of senior engineers who command premium rates.
This is not an incremental change. It is a structural break. The competitive landscape will be reordered from a pyramid of headcount to a hierarchy of AI leverage.
Here's the difficult part. I've spent the better part of two decades analyzing technology-driven cost structures, and I can tell you the 25% figure is not evenly distributed across service types. Code generation is far more automatable than legacy system maintenance. Testing can be heavily automated, but security audits still require human judgment. The average hides the variance.
The real story of IT services in the age of AI will be a story of the uneven distribution of cost reductions. A firm that promises 25% across the board may be setting itself up for failure when a client's specific project turns out to be one where only 8% is achievable.
The Contrarian Lens
Now, let me offer a perspective that most coverage of this news will miss.
Consider what this announcement says about the culture and psychology of the IT services industry. For three decades, these firms have been selling one thing: certainty. Certainty of delivery, certainty of staffing, certainty of price. AI erodes all three. When a CEO publicly announces that AI cuts costs by a quarter, he is telling his clients—and his own engineers—that the product is now cheaper to produce.
There is a deeply personal consequence here, one that the industry is not discussing. If AI reduces the need for labor, what happens to the millions of early-career engineers who entered this industry hoping for a stable career? India's IT industry is not just an economic engine; it is a social elevator. Every year, thousands of graduates from tier-two cities enter the industry through entry-level coding positions. These are the roles most vulnerable to automation.
And here is the paradox, which I've reflected on through the lens of my experience during the 2017 ICO boom and the 2020 DeFi liquidity crisis: every time the industry I follow becomes more "efficient," it becomes less inclusive. The middle-class aspiration that built the Indian IT engine—the dream of a better life through learning to code—is being traded for a marginal improvement in quarterly earnings.
Volatility is the tax on impatience, but this is not volatility. This is a structural replacement. And the cost is not measured in dollar terms alone.
The Execution Gap
Let's step back and evaluate what actually happens next. The 25% reduction is not a technological outcome. It is a management outcome. The technology is ready; the organizational capacity to absorb it is not.
I have watched this pattern before. During the ICO boom, I audited smart contracts that promised automation of everything from payments to governance. Most failed not because the code was bad, but because the organizations could not adapt to what the code required. The same will be true for IT services firms. The AI tools will work. The challenge is how to redesign a delivery model that was built around the assumption of human time.
This is a managerial problem, not a technical one. And for a mid-tier firm like Hexaware, the resources to solve it are scarce. The top-tier firms—Accenture with $30 billion in AI investments, Infosys with the Topaz platform, Wipro with its $1 billion commitment—are building at a pace that widens the gap with mid-tier players. AI-native firms, meanwhile, are attacking from below with targeted, automated services that require no human labor at all.
In between, Hexaware and its peers are the squeezed middle. Their response, reflected in the CEO's announcement, is to be the first to admit what everyone already knew: AI is the new cost structure. But admission is not adaptation. And adaptation, in this context, does not mean merely adopting AI tools. It means fundamentally redefining the firm's value proposition.
The Question of Pricing
The deepest problem is pricing. If we assume the 25% cost reduction is real, the value of a delivered project—in the eyes of the client—has not changed. It is worth the same amount as before. What the 25% represents is margin redistribution, not new value creation. The client, rightly, will demand a share.
The pricing model that emerges will not be "time per hour." It will be "value delivered." And value delivery, in this new world, will be measured by the speed and accuracy of the outcome. The firms that can price this way will capture the efficiency premium. The ones that cannot will be relegated to commodity status, competing on price alone.
This is the future that Hexaware's CEO is implying, but not articulating. The 25% is the first step toward a new pricing model. But no one knows how to price what does not yet have a market.
Takeaway
The 25% figure, repeated across the industry, will become a standard negotiating anchor. But the deeper truth is that the IT services industry is being forced to re-evaluate its very reason for being. It is no longer enough to deliver code. The future demands delivering outcomes.
And there is a question that the market has not yet answered: if the industry's largest asset—human capital—is no longer the primary source of value, what exactly is the firm selling?
We will know the answer in the next 24 to 36 months. The industry that sells "certainty" will be sold a new definition of certainty—one that is built not on the reliability of human labor, but on the predictability of machine-driven outcomes.
The 25% is the opening number in a larger negotiation. It is a negotiation that will determine not just the future of IT services, but the future of work for the millions who built it. And I would be lying if I said I know how it ends.
