By Srinath Sridharan, Author, corporate adviser & independent director on corporate boards | X: @ssmumbai
Digital systems are increasingly treated as elegant answers to problems that once demanded institutional reform, political negotiation, and administrative judgement. In conferences, boardrooms, and policy discussions, technology increasingly carries the aura once reserved for institutions.
The optimism rests on a powerful assumption. Digital systems would remove the intermediaries who once controlled access to markets and opportunity. For years, economists criticised middlemen for distorting exchange. The digital economy promised disintermediation, where platforms would connect citizens directly with customers, lenders, employers, and audiences.
What has emerged instead is a different structure of power. Markets are not becoming free of intermediaries; they are becoming dependent on new ones.
In digital markets, algorithms increasingly determine who meets whom. They decide which products appear before consumers, which job applicants survive the first filter, which borrowers appear creditworthy, and which voices travel across networks. These systems analyse vast data, predict behaviour, and allocate visibility at a scale no human intermediary could match.
What we are witnessing is the rise of algorithmic gatekeeping, where access to markets is filtered through code. The challenge for policymakers is no longer only regulating platforms, but understanding the algorithmic power embedded within them.
Consider the small entrepreneur building a business on an online marketplace. Customers from across the country begin discovering products through search results and the enterprise expands beyond its local geography. Then the pattern changes. Orders fade even though quality and ratings remain strong. The only shift lies in the ranking system that determines visibility. In that moment, the entrepreneur encounters the real gatekeeper of the digital marketplace. This marks a deeper shift in economic authority.
Efficiency does not automatically translate into fairness or transparency. When the mechanisms allocating opportunity become opaque, citizens lose the ability to understand how outcomes are produced. In many cases, these systems are owned elsewhere, reducing citizens and markets to paying participants in infrastructures they neither control nor shape.
The paradox deepens when one observes the confidence with which we speak about governing these systems.
Artificial intelligence panels multiply across media, business, and policy forums. Reports circulate describing ethical algorithms and responsible digital markets. Strategy documents promise frameworks that will balance innovation with accountability. The tone often suggests that institutions have already mastered the forces reshaping economic life.
The reality is less settled. Algorithmic systems combine advanced mathematics, behavioural modelling, and machine learning that evolves continuously with new data. Even their creators cannot fully predict outcomes. Institutional expertise rarely develops at the same pace as technological change. In many discussions about digital governance, the vocabulary of expertise travels faster than the expertise itself. Over-confidence in technological narratives often exceeds institutional exceeds the institutional capacity.
Into this uncertainty has emerged a new ecosystem of interpreters. Consulting firms and policy advisory groups increasingly operate at the intersection of governments and regulators, shaping how emerging technologies are understood and governed. Many perform valuable analytical work, yet they also occupy an unusual position in the digital economy. In effect, their counsel is often shaped by the unseen highest-paying voices in the room, not necessarily by the long-term imperatives of sovereign capability.
This dual layer of intermediation raises a deeper question of digital sovereignty. Control over data, platforms, and algorithmic capability is now strategic. India has built impressive digital infrastructure, but sovereignty cannot rest on infrastructure alone. Much of the architecture powering its digital life remains owned elsewhere, leaving the country rich in data generation but dependent in its interpretation and monetisation. Sovereignty, in that sense, remains more aspiration than reality.
History offers a familiar warning. Every technological revolution concentrates power before institutions learn to govern it. The deeper risk is not technological change itself, but the migration of authority into systems and actual owners, opaque to those they govern. The middleman of the digital economy increasingly resides in code, and in those who confidently interpret it on behalf of those who own it.
India’s technological ambition deserves admiration, but ambition alone does not create sovereignty. If the country wishes to lead the digital century, it must recognise that algorithms are no longer merely instruments of efficiency, but emerging gatekeepers of economic life.
Without that capability, a society risks celebrating technological progress while economic gatekeeping quietly migrates into systems that citizens do not understand and institutions often speak about with more confidence than comprehension. The enthusiasm surrounding emerging technologies can resemble a collective performance. For all we know, the next fashionable step may be to append “.ai” to surnames or corporate identities simply to signal modernity, much as companies once rushed to adopt “.com”.
When deeper questions of capability, sovereignty, and control arise, they are often drowned out by digital cheerleading and amplified narratives that money can manufacture. That performance may hold for a while, but sovereignty cannot be built on branding or borrowed confidence. It demands introspection, institutional competence, and the rebuilding of a scientific and technological base worthy of a country that seeks to shape its own future. A nation that outsources the engines of its digital future will eventually discover that sovereignty cannot be downloaded later.