Beyond productivity: rethinking how we invest in AI

When Zoho founder Sridhar Vembu recently warned that the rapid adoption of artificial intelligence could eventually create a purchasing power problem by reducing entry-level hiring, the reaction was predictable. The discussion quickly turned into a familiar debate over whether AI would replace software engineers and eliminate jobs.

Yet that framing understates the significance of his intervention. Vembu’s remarks are not merely about employment. They are about how organizations and governments are making decisions on one of the most consequential technologies of our time.

The question is no longer whether companies should invest in AI. They should. Artificial intelligence has the potential to transform productivity, improve customer experience, accelerate innovation and create entirely new business models.

The real question is whether these investments are being evaluated with the same strategic discipline that organizations apply to other major capital decisions.

Unlike a factory, an acquisition or a research and development program, an AI investment is likely to be celebrated simply because it exists. Success is measured by lower costs, fewer employees and higher productivity.

Those are undoubtedly important outcomes. But they are also incomplete. As AI becomes central to corporate strategy, organizations need a broader framework for evaluating what these investments actually create, and what they may unintentionally erode.

AI investments need better success metrics

Every significant corporate investment is judged through multiple lenses. A new manufacturing plant is expected to improve capacity, strengthen supply chains and generate future growth. Research and development expenditure is assessed not only by immediate returns but by its ability to create intellectual property and long-term competitive advantage. Human capital investments are justified because they build organisational capability over time.

AI deserves the same treatment.

Today, boards often justify AI expenditure through familiar financial metrics: productivity gains, cost savings, operating margins and shareholder returns. While these are legitimate objectives, they represent only part of the value AI can generate. A narrow focus on efficiency risks encouraging organizations to deploy AI primarily as a cost-cutting tool rather than as a capability-building technology.

Boards should therefore ask broader questions. Has AI improved decision-making? Has it enabled employees to innovate faster? Has customer experience improved? Has it strengthened organizational learning? Has it created new products, services or business models?

Most importantly, has it enhanced the firm’s long-term competitive capability rather than merely improving the next quarter’s earnings?

The companies that derive the greatest value from AI are unlikely to be those that simply automate existing work. They will be those that redesign how work is performed and how value is created. AI should therefore be viewed as a strategic investment in organizational capability, not merely an exercise in operational efficiency.

The apprentice gap: Who builds tomorrow’s experts?

Perhaps the least discussed consequence of AI adoption is its effect on how expertise is developed.

Every knowledge profession relies on apprenticeship. Doctors begin with routine clinical work before making life-and-death decisions independently. Lawyers spend years researching and drafting before arguing landmark cases. Academics build expertise through years of research before becoming recognized scholars. Software engineers similarly learn by debugging code, fixing errors, reviewing systems and gradually assuming greater responsibility.

These routine tasks are often portrayed as repetitive work that AI can easily automate. But they also constitute the training ground where professional judgement is formed.

If AI increasingly performs entry-level work while organizations fail to redesign how expertise is cultivated, companies may inadvertently weaken the pipeline that produces future architects, engineering managers, product leaders and chief technology officers. The savings realised today may come at the cost of organizational capability tomorrow.

This is not simply a labor-market issue. It is a knowledge-management challenge.

Businesses have always understood the importance of succession planning for leadership. AI demands a similar conversation about succession planning for expertise. Organisations must ask not only how AI replaces tasks, but also how future professionals will acquire the judgement, intuition and contextual understanding that cannot be downloaded from a model or generated through a prompt.

The AI race needs strategic discipline, not herd behavior

There is little doubt that AI represents a transformative technological shift. Yet history also reminds us that transformative technologies often create waves of imitation alongside genuine innovation.

Organizations sometimes adopt new technologies because they solve real business problems; they also adopt them because competitors, consultants and investors expect them to. AI risks creating a similar dynamic.

Today, announcing an AI initiative often signals that a company is technologically progressive. But signaling should not be mistaken for strategy. The real question is whether AI is genuinely transforming workflows or merely being inserted into existing processes to satisfy market expectations.

Boards should therefore apply the same rigor to AI investments that they would apply to any other strategic decision. Which workflow is being improved? Which customer problem is being solved? What measurable capability has been created? Would the investment still make sense if competitors were not making similar announcements?

These questions are particularly important for India. Much of the global AI conversation is shaped by economies facing aging populations and high labor costs, where replacing labour is often commercially rational. India’s challenge is different. Its comparative advantage has long been its abundant pool of skilled human capital.

The objective, therefore, should not simply be to substitute workers with algorithms, but to use AI to amplify the productivity, creativity and global competitiveness of India’s workforce. The distinction is subtle, but it carries profound implications for how firms invest and how policymakers think about technological transformation.

The same strategic discipline should extend to public policy. Governments should not regulate AI adoption by requiring companies to justify every innovation against employment targets. Such an approach would risk slowing technological progress. They should, however, develop far stronger labor-market intelligence to understand how AI is reshaping occupations, skills and wages.

Just as governments monitor inflation, industrial production or financial stability, they should continuously assess how AI is transforming the world of work so that education systems, training programs and labor policies evolve alongside technological change.

The debate sparked by Sridhar Vembu should therefore not end with predictions about job losses. Artificial intelligence undoubtedly needs investment, ambitious budgets and rapid innovation. But it also needs thoughtful management.

The most successful organizations and economies will not be those that simply spend the most on AI. They will be those that invest with strategic clarity, measuring success not only by the productivity AI delivers today but by the capabilities, expertise and resilience it helps build for tomorrow.

Anu Singh Lather is vice chancellor of Dr. B.R. Ambedkar University, Delhi. Tarun Agarwal, PhD., is an associate fellow at the Center of Policy Research and Governance, New Delhi.