OPINION | India AI at Inflection Point: Capital, code and consequence

AI systems are no longer curiosities.

In a prominent annual investor conference held abroad recently, a revealing trend emerged. These conferences are ritual marketplaces of reputation. Companies arrive hoping to attract celebrated fund managers. Fund managers arrive seeking private audiences with prospective investees. The list of the ten most sought-after firms usually becomes a barometer of emerging opportunity. This time, an entire category was missing – Indian companies.

Absence is sometimes more eloquent than presence. Consider what foreign portfolio investors have been doing in recent months. In December 2025 alone, overseas investors withdrew over ₹22,600 crores from Indian equities and nearly ₹15,000 crore from Indian debt. For the full calendar year, equity outflows touched an unsettling ₹1,66,000 crore. Capital does not migrate at this scale without a reason.

The reason is becoming harder to ignore. India does not yet possess a persuasive Artificial Intelligence story to tell the world.

This matters far beyond the narrow anxieties of one sector. For three decades, India’s software services industry offered a simple proposition. Import global tools and frameworks. Train vast pools of engineers to use them. Execute custom projects for international clients at competitive prices. The advantages were clear: abundant manpower, process discipline and favourable currency conversion.

That era is over. Artificial Intelligence is compressing complexity, accelerating execution and reducing the economic value of routine labour. The very foundation of India’s technology model is under strain.

The Disruption of Value Chains

Around the world, AI systems are no longer curiosities. They are active participants in corporate supply chains. Large language models (LLMs) can generate functional code from plain prompts. They can summarise dense documents. They can classify unstructured data. They can convert raw information into usable signals. Tasks that once demanded weeks of human effort are increasingly completed in minutes.

The transformation is visible in business process management, customer support, data analytics and software development itself.

* First-level help desks are being replaced by intelligent digital assistants.

* Quality assurance roles are shrinking as AI performs automated testing.

* Knowledge processing operations are discovering that machines can draft reports, extract insights and prepare presentations without fatigue or resentment.

Every layer of intermediation that depended on lower skilled or experienced or lower-hierarchy talent is being challenged by digital technologies, available commercially.

Take the example of Security Operations Center as a Service, a field familiar to technology professionals. Such services collect metadata from thousands of machines, endpoints and users within an enterprise. They normalise this data into common formats. They apply rules to convert data into risk signals. Teams of analysts then examine those signals in tiered structures and transform them into actionable intelligence. With AI, each of these steps has been radically accelerated. Data normalisation that demanded specialist teams is now handled by algorithms. Rule creation that required patient experience is being drafted by models. Analyst layers are thinning as agent-based AI interfaces perform triage and investigation.

What is true of this illustration is increasingly true of dozens of others. Artificial Intelligence does not merely improve individual tasks. It reorganises entire industries. It rewards those who design new systems from first principles. It punishes those who attempt timid modifications to old processes. Such a global conference list is an indicator on where future value will be captured.

The Institutional Test Before India

For Indian firms, the challenge is uncomfortable because it strikes at familiar habits. Many listed service companies have long preferred margin protection over long-gestation research. The culture of incrementalism served them well when technology changed slowly. It is inadequate when technology changes at the speed of imagination.

The more serious vulnerability lies beyond companies and extends to institutions. Boards of directors, regulators and political leaders must recognise that AI transition is not a technical upgrade. Capital allocation decisions will determine which enterprises survive and which become footnotes. Workforce policies will determine whether millions of mid-career engineers are re-skilled or rendered obsolete. Regulatory frameworks will determine whether India becomes a consumer of imported intelligence or a creator of global platforms.

Today, global investors are attracted to economies that demonstrate credible commitments to AI innovation. They look for compute infrastructure, deep research ecosystems, predictable data governance regimes and companies willing to invest patiently in new capabilities. On each of these dimensions India is only beginning its journey.

The temptation for protectionism will inevitably rise. Under pressure from industry associations, some firms may urge the government to reserve sensitive technology contracts only for domestic entities, invoking national security and geopolitical volatility. Such strategies might offer temporary comfort. But they risk producing an insular marketplace where Indian models work only within Indian boundaries and remain incompatible elsewhere. That would be a strategic tragedy disguised as tactical success.

The true way out demands courage. India must invest in sovereign compute capacity and world-class AI research clusters. Universities must be funded to move from routine IT training to foundational AI development. Listed companies must be encouraged through incentives and disclosures to articulate their AI transformation roadmaps. Regulators must craft frameworks that combine innovation with safety so that global clients and investors trust Indian systems. Boards must induct directors who understand the language of code and consequence. None of this can be achieved through slogans. It requires institutions willing to think in decades rather than quarters.

From the First Curve to the Second

The next phase of India’s growth will belong to those who accept the new paradigm. Services must be designed with AI embedded into processes from the beginning rather than bolted on as an afterthought. Firms that learn to blend human judgement with machine intelligence could yet emerge as competitive champions. They may be few at first. But they will carry disproportionate influence on investor imagination.

For the rest, the future is predictable. They will struggle to win new contracts as agile AI-native companies outcompete them on features and cost.

India stands therefore at a rare national crossroads. The country has crossed the first quarter of the 21st century with confidence and momentum. But the remaining three quarters will be defined by choices being made now in cloud rooms, classrooms, boardrooms and cabinet meetings. Artificial Intelligence is becoming the organising principle of global economics.

Polity leaders must read the signals clearly. This is not only about one conference or one month of outflows. It is about whether India works in becoming an architect of the AI century.

(Srinath Sridharan is a corporate advisor and author of ‘Family and Dhanda’, Anand Venkatanarayanan is strategic security and digital policy researcher.)

Views are personal and do not represent the stand of this publication.