AI can make India’s social security portable

The promise is of efficient tracking, ensuring contributions and benefits can move with the worker across districts and states

The operationalisation of the Social Security Code, followed by draft rules placed in the public domain, marks a clear statement of intent in how India seeks to safeguard its workforce. By bringing gig workers, platform labourers and the unorganised sector under a single framework, the law recognises how work actually happens today and places the worker, not the contract, at the centre of policy. The intent is unambiguous. The task now is delivery: converting this opportunity into protection that workers can see, access, and rely on.

The scale of that task is evident. Nearly 90 per cent of India’s workforce continues to operate in informal employment arrangements. At the same time, social security coverage has expanded steadily over the past decade, rising from about 19 per cent of the population in 2014 to nearly 64 per cent today, according to ILOSTAT’s SDG indicators. In absolute terms, this translates to roughly 94 crore beneficiaries, placing India among the world’s largest social security systems. These figures point to institutional capacity already operating at a population scale.

What gives this expansion its real significance is its operational character. Unlike survey-based estimates, India’s social security architecture is anchored in live administrative systems. Organisations such as EPFO and ESIC process crores of contribution, compliance, and benefit transactions every day across establishments, sectors, and states, generating real-time records of employment, wages, and protection. Together, these systems reflect how labour markets function daily.

The availability of such live data changes what policy delivery can realistically aim for. When employment spells, wages, and contributions are recorded in real time, social security need not remain static; it can follow the worker as circumstances change. This is where the promise of the Social Security Code moves from legal design to administrative possibility. That promise must contend with the defining feature of India’s labour market: mobility. Workers routinely move between employers, platforms, occupations, and states, often combining multiple forms of work within a single year. A social security system designed around the worker, rather than a single employer or occupation, aligns naturally with this reality.

Because India’s administrative systems have evolved in silos, shaped by the legacy of scheme-specific delivery, worker identities and contribution histories often reset as individuals move across schemes or jurisdictions. The moment now calls for breaking this historical baggage. Instruments such as the Universal Account Number under EPFO, already extended into platforms like e-Shram, provide a foundation for continuity, allowing employment histories and entitlements to travel with the worker rather than remain confined within institutional boundaries.

Seen this way, the nature of the challenge changes. India has already demonstrated that it can scale social protection. The design imperative now is portability: ensuring coverage, contributions, and benefits move seamlessly with the worker, without repeated administrative resets that raise transaction costs and risk exclusion.

It is at this stage that artificial intelligence becomes relevant as a governance capability. The value of AI lies not in prediction alone, but in connection. The social security administration already generates vast volumes of transactional data across states, tracking employment spells, wage bases, and contribution histories at a scale few countries can match. When systems operate independently, continuity breaks down as workers move. When they are connected intelligently, continuity becomes possible.

Embedded into routine administration, AI can help identify employment transitions, flag contribution lapses, and highlight eligibility gaps before exclusion occurs. It enables risk-based compliance by allowing inspectors to focus on patterns rather than paperwork, while improving responsiveness without increasing procedural burden. Used this way, AI strengthens institutions rather than sitting outside them.

This institutional role for AI fits naturally within India’s federal structure. With labour on the Concurrent List, the Union and states are delivery partners. States remain closest to workers and employers, shaping inspections, enforcement priorities, and welfare delivery. National digital systems provide common foundations through shared identifiers and interoperable standards that allow benefits and contributions to move with workers across district and state boundaries. In practice, this strengthens State capacity without centralising control.

Proceed with caution

AI, however, is not a shortcut to universal social security. Earlier phases of digitisation offer clear lessons. The first is to avoid digitising poorly designed processes. AI layered over unclear rules or weak data will amplify existing inefficiencies. Business process redesign, clarity in rules, and data cleaning must therefore come first. The second lesson is the need to move beyond administrative silos, since AI draws its value from patterns across systems.

Sequencing matters. Initial efforts should focus on low-risk, high-impact use cases such as cleaning registries, clearing backlogs, detecting compliance gaps, and reducing repetitive administrative work. Once these basics are in place, sandboxing offers a bridge from design to scale. Platforms such as AIKosh support responsible experimentation, allowing solutions to be tested and refined before wider deployment.

Trust remains the thread that holds the system together. Employers need predictability. Workers and unions expect transparency. State governments must see national systems as enabling rather than intrusive. AI systems built with shared dashboards, common standards, and clear audit trails can reduce discretion and strengthen trust across stakeholders.

At the G20 in New Delhi, Prime Minister Narendra Modi articulated a broader vision of digital public infrastructure and AI as tools for public good, aimed at expanding access, building trust and strengthening State capacity. This vision aligns with the Sustainable Development Goals on universal social protection, decent work, and reduced inequality. For India, these commitments converge with the aspiration of Viksit Bharat, where economic growth is matched by inclusion.

The Social Security Code offers India a new promise, one that reflects how people work today across jobs, platforms, and states. Used well, AI can make governance not just efficient, but deliberately humane. Anchored in institutional strength, technology turns execution into reform. That is how India can show the way in delivering public policy at scale.

(Uttam is Regional Provident Fund Commissioner [Kochi and
Lakshadweep]; Srinivas is a faculty at IIM Visakhapatnam; Jay is Scientist-E at the Ministry of Electronics and Information Technology)

(Disclaimer: The views expressed above are the author's own. They do not necessarily reflect the views of DH.)