India’s Next Literacy Divide: Who Can Question the Machine?

​A small manufacturer can now generate a polished business proposal in minutes. A student can produce an impressive assignment without understanding its argument. An artisan can describe a product in a language never spoken at home. These are useful possibilities. They also reveal a difficult question. When a machine makes everyone sound knowledgeable, how does society distinguish confidence from competence?

India’s AI future will depend partly on its ability to answer this question. Access to artificial intelligence can spread much faster than the ability to judge it. That gap could become a new source of economic inequality.

A Canadian announcement raises an Indian question. On September 9, 2026, Canada launched a National AI Literacy Initiative with the Alberta Machine Intelligence Institute. Its three learning streams cover students, educators and Canadians more broadly, with an emphasis on practical understanding and responsible use. The significance lies in treating AI literacy as a public capability that must reach beyond technology professionals. (canada.ca)

India has already taken steps in this direction. In November 2025, the government launched YUVA AI for ALL, a free, 4.5-hour foundational course under the IndiaAI Mission, with an ambition to reach one crore citizens. It covers basic understanding, Indian applications and responsible use. That figure is a target, however, and should not be confused with demonstrated capability among one crore people. (pib.gov.in)

The policy question therefore moves beyond introducing another course. What happens when the learner closes the screen and returns to a classroom, a workshop, a shop counter or a village enterprise? Does the learning survive contact with actual work?

History warns against confusing equipment with capability. Industrial progress has always required institutions that help people absorb technology. Machines needed operators, repair services, reliable power and production discipline. Computers needed changes in record keeping and management. Agricultural technologies needed practical advice suited to local conditions. The value emerged through the system around the technology.

AI will face the same test. A business with incomplete cost records cannot assume that an intelligent tool will produce a reliable quotation. A factory that does not record why products are rejected cannot expect software to discover every cause. Some businesses may first need consistent records and clearer processes. Calling these basic improvements an AI transformation will not make the underlying work disappear.

India’s challenge is especially demanding because enterprises differ enormously in size, language, management capacity and access to advice. A common introductory course can create awareness. Applying that awareness requires much more local understanding.

The smallest business faces the largest translation problem. Consider a garment unit in Tirupur, an engineering workshop in Ludhiana or a handicraft enterprise in Jaipur. The owner is likely to judge technology through immediate concerns: preparing quotations, responding to buyers, reducing mistakes, understanding specifications and getting paid.

Training should begin with those tasks. Participants could prepare a buyer response, check it against the actual order and identify what the system has invented. They could draft a product description and remove claims that the product cannot support. They could compare an AI-generated costing sheet with verified inputs and discover why a convincing answer can still be commercially dangerous.

This would teach both usefulness and judgement. A demonstration that produces an attractive answer teaches only half the lesson.

The IndiaAI Mission has an approved outlay of ₹10,371.92 crore and includes skills, applications, computing capacity and safe and trusted AI among its pillars. The government has explicitly linked its objectives to MSME productivity and competitiveness. These are substantial commitments, but expenditure and programme coverage cannot by themselves establish whether a small firm has become more capable. (pib.gov.in)

The next divide may run through the ability to verify. An experienced professional can often detect an implausible answer because years of work provide a reference point. A beginner may have no such protection. Both can access the same tool, yet the consequences can be very different.

This creates an uncomfortable possibility. AI could widen the advantage of people who already possess knowledge, networks and access to expert review. Those with fewer resources may rely most heavily on answers they are least equipped to check.

AI literacy must therefore include the confidence to challenge the output. Where did this information come from? Is it current? What has been assumed? What needs independent confirmation? Which information should never have been entered into the system?

These questions belong in everyday education. They are part of economic self-defence.

Clusters could become the practical classrooms. India’s industrial clusters offer a useful setting because neighbouring enterprises often face related problems. A shared learning programme can work with familiar products, documents and production situations. Participants can see whether an application helps a business resembling their own.

Industry associations, polytechnics, ITIs, enterprise development organisations and common facility centres could jointly organise these programmes. Technical specialists would validate the AI content. Local advisers would explain the business setting. Experienced workers would contribute the knowledge that formal training often overlooks.

A cluster could begin with a small group of enterprises, select two or three manageable tasks and provide follow-up support. Progress could be assessed through verified changes in task time, errors, quality and participant understanding. Findings should include applications that proved unsuitable. Public programmes need the freedom to report that a tool added little value.

Such an approach would also create a meaningful role for business associations. Their contribution could extend from organising seminars to maintaining a continuing service that helps members learn, test and make informed decisions.

Workers need a voice in the transition. A programme focused entirely on owners may miss the people who understand how work actually happens. Machine operators, supervisors, accounts assistants and sales staff often know where delays and errors originate.

Their involvement also changes the purpose of training. If AI saves time, what happens to that time? Does work become safer and more skilled? Do employees gain opportunities to learn? Or does the same workforce simply face tighter targets?

These outcomes will depend on management choices and institutional arrangements. Training cannot guarantee good employment, but it should make workers more capable of participating in those choices. Paid learning time, accessible instruction and opportunities to question new systems would make that commitment more credible.

The future requires public judgement as well as private innovation. As AI becomes embedded in software and services, people may encounter it without deliberately choosing an AI tool. Literacy will then mean understanding when an automated recommendation deserves attention, when it requires checking and how to seek human review.

India should prepare for that future through regional-language instruction, practical examples and trusted local support. Translation alone will be insufficient. A fluent answer in a familiar language can still be wrong. Accessibility must be accompanied by the ability to assess reliability.

Canada’s initiative offers a useful prompt for India to examine the distance between national ambition and everyday capability. India already has programmes on which to build. The next task is to connect learning with the places where people produce, trade, teach and earn.

The decisive test will come when a small entrepreneur encounters a confident but incorrect answer and has the knowledge to reject it. A country becomes more capable when its people can use powerful tools without surrendering their judgement.

​The Garment Must Now Carry Its History

Why proof of origin could become the next dividing line in India’s textile exports

A garment can have perfect stitching, the right colour and an attractive price, yet still face questions that its quality cannot answer. Where did the fabric come from? Which factory processed it? What actually happened in India? Can the exporter connect the finished product to the materials and production stages behind it?

For India’s textile industry, these questions reach far beyond the shipping department. They suggest a future in which manufacturing competitiveness will depend partly on the ability to prove how manufacturing happened. The fabric will carry the design. The records will carry its credibility.

The return of geography. For decades, international trade encouraged businesses to divide production across countries. Cotton could originate in one economy, become yarn in another, turn into fabric elsewhere and finally reach a garment factory thousands of kilometres away. This was the working logic of global production.

The removal of restrictions under the WTO Agreement on Textiles and Clothing on 1 January 2005 marked a major historical transition. Textile trade moved out of its special quota regime and into the general multilateral trading framework. Geography continued to matter, but firms gained greater room to organise production around costs, capabilities and delivery. (wto.org)

Today, geography is acquiring a different commercial meaning. The country attached to a product can shape its treatment at the border and the scrutiny it attracts. The emerging contradiction is sharp: businesses built international supply chains to become efficient, while governments increasingly demand a clearly established national origin for the final product.

On 3 September 2026, Commerce Minister Piyush Goyal called for credible Indian country-of-origin certificates and genuine value addition, warning against India becoming a channel for goods from elsewhere through marginal processing. His message places the credibility of Indian production at the centre of export policy. (indianexpress.com)

Suspicion must not become evidence. The White House report on transshipment places India among large trading economies where suspected risks exist alongside substantial legitimate commerce. However, it also acknowledges that changing trade patterns do not establish that displaced Chinese exports were illegally rerouted. Real investment, production relocation and legitimate sourcing changes explain part of the shift. (whitehouse.gov)

That distinction deserves more attention than the accusation. An increase in exports is a reason to examine production evidence, not a verdict against the exporter. If every change in sourcing becomes suspicious, countries are effectively being discouraged from achieving the supply-chain diversification that international buyers have encouraged.

India therefore needs a firm position on both sides: false origin claims should face scrutiny, and legitimate manufacturers should be protected from broad allegations unsupported by shipment-level evidence. Export credibility requires honest businesses and fair enforcement.

The small supplier becomes the large exporter’s vulnerability. Consider an illustrative garment order passing through a fabric supplier, a dyeing unit, a cutting contractor, an embroidery workshop and a stitching factory. The finished product may be entirely consistent with the declared production process. Yet the evidence may sit in disconnected invoices, handwritten job-work registers and messages on different phones.

The difficulty is not simply having documents. It is connecting them. A purchase invoice identifies a transaction, but may not establish which fabric batch entered a particular export order. A production entry records activity, but becomes more useful when it can be linked to material quantities and dispatch records.

Origin documentation also serves a different purpose from labour and environmental evidence. A record showing where fabric was processed does not establish safe working conditions or responsible wastewater treatment. Exporters need related systems that can answer these different questions without pretending that one certificate proves everything.

The commercial risk is that a weak link in documentation can spread beyond the supplier responsible for it. An exporter may face questions it cannot answer quickly, even when the underlying production is legitimate.

The hidden price of proving innocence. Documentation has an economic cost. It takes staff time, training and coordination. When records are questioned, delayed acceptance or payment can add financing pressure.

Consider a purely illustrative calculation. An additional compliance cost of ₹50,000 equals 2.5% of a ₹20 lakh order, but only 0.25% of a ₹2 crore order. The same fixed cost creates very different burdens. This is why apparently uniform requirements can favour larger firms.

A future textile industry could therefore become more concentrated even without major differences in manufacturing efficiency. Small producers may lose access because they cannot afford the systems needed to demonstrate their capabilities. Policy should recognise this possibility before interpreting every business exit as a failure of productivity.

Clusters need a shared capacity to establish facts. Tiruppur, Surat, Ludhiana, Panipat and Jaipur should treat traceability as a collective development challenge. Shared testing facilities help firms establish product quality. Shared documentation services could help them establish the production history behind each shipment.

The starting point should be practical: consistent supplier declarations, clear material lists, batch references and job-work records that connect inputs to finished goods. Cluster associations could support trained documentation teams and affordable digital tools, with independent checks where appropriate. Such services would support compliance; they would not replace the applicable origin rules or guarantee customs acceptance.

The system should accommodate the small workshop that contributes specialised work but has limited administrative capacity. Simple interfaces, local-language assistance and reusable records would matter more than an impressive technology launch.

Confidentiality also matters. Suppliers should not have to expose their entire customer base, pricing or commercial relationships to every participant. Shared infrastructure needs clear access controls and accountability.

Digitisation alone will solve little. A false declaration remains false after it enters a database. The real task is to make records consistent with actual production.

Predictability must travel in both directions. Exporters are being asked to provide greater certainty about their supply chains. They need greater certainty about domestic policy as well. The official RoSCTL extension covers apparel and made-ups until 30 September 2026, or approval for the next Finance Commission cycle, whichever comes earlier. 

This uncertainty affects pricing decisions. Businesses accepting orders for later delivery need to understand the support available when those shipments leave. Demanding precise commercial planning while leaving policy decisions close to expiry creates an avoidable burden.

Origin integrity should therefore be strengthened through standard records, targeted checks, clear explanations and timely resolution of disputes. Repeatedly asking every small exporter to reconstruct the same production history would consume resources without necessarily improving enforcement.

The next export advantage will include the cost of being believed. In the coming years, buyers and border authorities may become better at comparing shipment records, supplier relationships and claimed production capacity. Exporters with coherent evidence could become easier to assess and more attractive to buyers seeking dependable supply.

India has an opportunity to make that confidence accessible across its textile clusters. The strategic challenge is to help thousands of small firms demonstrate real production without burying them under the cost of doing so.

Made in India must carry a history that can be checked. The test of good policy will be whether an honest small manufacturer can tell that history clearly, affordably and once.

#TextileExports #MSMEs #MadeInIndia #Traceability #ClusterDevelopment

Why Tax Collection Does Not Reveal the Real Industrial Map of India

One Country, Three Different Economic Maps

India does not have one economic map. It has at least three. The first shows where goods are manufactured. The second shows where income and value are created. The third shows where taxable transactions are recorded. These maps overlap, but they are not identical.

This difference becomes visible when GST collections are compared with industrial production. Maharashtra leads GST collection because it combines manufacturing, finance, corporate headquarters, ports, services and a large consumer market. Gujarat has a far more industry-intensive economy, but its GST share is lower than Maharashtra. Delhi produces relatively little industrial output but collects substantial GST. Haryana contributes only around 3.6 to 3.7 per cent of India’s GDP, yet it generates approximately 7.1 per cent of the domestic GST attributed to states.

The simple conclusion would be that Haryana is producing far more than its economic size suggests. The more accurate conclusion is different. Haryana has become one of India’s densest centres of formal, taxable and corporate economic activity. Its GST strength comes not only from factories, but also from Gurugram’s corporate economy, NCR consumption, automobile trade, warehousing, logistics, real estate and business services.

This is why GST should never be treated as a direct measure of industrialisation.

India’s Top GST States and Their Industrial Reality

During 2024–25, Maharashtra collected approximately ₹3.58 lakh crore in gross GST and contributed more than one-fifth of domestic GST attributed to states. Karnataka followed with around ₹1.59 lakh crore, Gujarat with ₹1.36 lakh crore, Tamil Nadu with ₹1.31 lakh crore and Haryana with ₹1.19 lakh crore. Uttar Pradesh, Delhi, West Bengal, Telangana and Odisha completed the leading group.

But their industrial structures are sharply different.

Industry contributes around 42 to 43 per cent of Gujarat’s state value added. In Odisha, the proportion is also above 43 per cent because mining, metals, power and large mineral-based industries dominate the economy. Tamil Nadu has a more diversified industrial structure, with industry contributing roughly one-third of its state value added. Maharashtra’s industrial share is only around one-fourth, yet it leads India in GST because it combines industrial production with finance, services, imports, consumption and corporate transactions.

Delhi represents the opposite extreme. Industry has a relatively small presence in its economy, but the city records high GST because it is a major centre of consumption, trade, distribution, professional services and company registrations.

The comparison reveals a fundamental fact. A state can be highly industrialised without becoming a proportionately large GST collector. It can also generate high GST without being a major industrial producer.

The Haryana Puzzle

Haryana is perhaps the most important case in this comparison.

Industry accounted for approximately 29.2 per cent of Haryana’s Gross State Value Added in 2023–24. Manufacturing contributed 17.7 per cent, construction 9.2 per cent, electricity and utilities 2.1 per cent, and mining only 0.2 per cent. Services contributed 52.9 per cent, while agriculture and allied activities accounted for 17.9 per cent. (ncaer.org⁠)

Haryana, therefore, is not overwhelmingly industrial in the way Gujarat or Odisha is. Its industrial share is close to the average of Indian states. Yet Haryana collected approximately ₹1,19,362 crore in gross GST during 2024–25 and ranked fifth among all states. Its per-capita GST collection was approximately ₹47,083, the highest among major states. (cdnbbsr.s3waas.gov.in⁠)

The contrast is striking. Haryana contributes approximately 3.6 to 3.7 per cent of India’s GDP, around 4.5 per cent of national industrial output and nearly 7.1 per cent of domestic GST attributed to states.

This is not an accounting accident. It reflects the unusual economic geography of the state.

Gurugram hosts the headquarters, regional offices and service operations of major automobile, technology, consulting, financial, real-estate and consumer companies. Faridabad, Manesar, Gurugram, Sonipat, Panipat, Yamunanagar, Bahadurgarh and other industrial centres add manufacturing depth. The state also benefits from proximity to Delhi, high household incomes, extensive road connectivity and a large formal business base.

Haryana is therefore more than an industrial economy. It is a manufacturing, consumption, logistics and corporate-registration economy operating within the wider National Capital Region.

GST Measures Transactions, Not Factories

The misunderstanding begins with the assumption that high GST collection must mean high production. GST is a destination-based tax. It broadly follows consumption and taxable transactions rather than the physical location of production.

A factory may manufacture a product in one state, but the final tax revenue can move towards the state where that product is consumed. Similarly, a corporate office, warehouse, service centre or large distributor can generate substantial GST without owning a large manufacturing plant.

GST collection is also influenced by formalisation. Two states may have similar levels of economic activity, but the state with better invoicing, stronger compliance, more organised retail and a larger registered business base may report much higher GST.

This creates an invisible divide between the formal economy and the productive economy. GST sees the part of economic activity that enters the tax network. It does not fully capture informal manufacturing, household enterprises, agricultural activity, exempt goods or the real technological quality of production.

A state can therefore collect high GST while having weak manufacturing capability. Another can produce large quantities of industrial goods but collect less GST because much of the final consumption occurs elsewhere.

The Historical Change from Production Centres to Transaction Centres

Before economic liberalisation, the industrial importance of a state was largely associated with factories, public enterprises, electricity generation, mining and physical infrastructure. Industrial maps were built around steel plants, textile mills, engineering centres, ports and mineral belts.

The post-1991 economy changed this relationship. Services expanded, supply chains fragmented and corporate functions became geographically separable from production. A factory could be located in one state, its head office in another, its warehouse in a third and its consumers across the country.

GST deepened this transformation after 2017 by creating a national indirect-tax system based largely on destination and invoice trails. It improved transparency and reduced many internal tax barriers, but it also made state-level GST collection a hybrid indicator. It now reflects consumption, formalisation, corporate organisation, logistics and services alongside production.

The modern economic centre is no longer always the place where machines are installed. It may be the place where orders are processed, invoices are raised, services are supplied, goods are distributed and final consumption takes place.

Why Gujarat and Odisha Look Different from Haryana

Gujarat’s industrial strength is rooted in chemicals, petrochemicals, engineering, pharmaceuticals, automobiles, textiles, ceramics, ports and energy-intensive production. Its industrial share is very high, and its contribution to national industrial output is estimated to be around 14 per cent.

Odisha’s economy is even more industry-intensive in proportional terms, but much of its industrial base is concentrated in mining, metals and capital-intensive production. These sectors can generate enormous output without producing an equally large number of taxable retail transactions or jobs.

Haryana’s industrial share is lower, but its economy produces more transactions per unit of output. Its location beside Delhi, high-income consumers, formal enterprises and corporate concentration enlarge the GST base.

This means Gujarat may be more industrially deep, Odisha more resource-intensive and Haryana more transaction-intensive. GST alone cannot reveal these differences.

The Danger of Rewarding Collection Instead of Capability

Policy can become distorted when high GST is treated as evidence of successful industrialisation. A state may improve tax administration and consumption without developing technological capability, industrial employment or domestic supply chains.

Similarly, a state with mines, power plants, steel factories and intermediate-goods industries may contribute significantly to national production but appear fiscally weaker because final demand and corporate transactions are recorded elsewhere.

The danger is that governments may begin competing mainly for headquarters, warehouses, commercial registrations and high-income consumption. These activities are valuable, but they cannot substitute for industrial capacity.

Factories create production ecosystems. They support tool rooms, repair services, logistics companies, component manufacturers, testing laboratories and skilled employment. Corporate offices create high-value jobs but often generate fewer backward linkages with local MSMEs. A balanced state economy needs both.

What Haryana Must Do Next

Haryana’s high GST performance is an advantage, but it can also hide structural weaknesses.

The state remains heavily concentrated around the NCR belt. Gurugram and Faridabad account for a disproportionate share of formal economic activity, while several districts remain dependent on agriculture, traditional industries or low-productivity services. This creates a state with world-class corporate zones existing beside regions with limited industrial diversification.

The next phase of Haryana’s development cannot depend only on real estate, automobiles, corporate services and NCR consumption. It must spread industrial capability into secondary cities and existing clusters.

Panipat can move from conventional textiles towards technical textiles, recycling and sustainable processing. Faridabad can deepen precision engineering, machinery and industrial automation. Manesar and Gurugram can expand from automobile assembly and corporate services into electric mobility, electronics, software-integrated manufacturing and advanced components. Sonipat can become stronger in food processing, logistics and consumer manufacturing. Ambala’s scientific-instrument cluster can move towards medical devices and precision technologies. Yamunanagar can modernise its plywood, paper and engineering base.

The objective should not merely be to increase the number of factories. It should be to raise local value addition, technology absorption, supplier capability and industrial wages.

A Better Way to Judge State Performance

States should be evaluated through a combined economic scorecard rather than a single number. GST collection must be read alongside manufacturing GSVA, total industrial output, exports, factory employment, industrial wages, electricity consumption, investment, technological intensity, MSME productivity and regional distribution.

Under such a framework, Maharashtra would emerge as India’s broadest economic platform. Gujarat would stand out for industrial depth. Tamil Nadu would be recognised for diversified manufacturing and employment. Karnataka would lead in high-value services and technology-linked production. Odisha would appear strong in resource-based industry but weaker in diversification. Haryana would emerge as a compact, formal and transaction-dense economy with substantial but geographically concentrated industrial capacity.

Such a comparison would be more honest than simply ranking states by GST.

The Future Economic Map

The states that dominate GST today may not automatically dominate industrial production tomorrow. The future will be shaped by electronics, semiconductors, electric mobility, renewable-energy equipment, advanced materials, defence production, biotechnology, data infrastructure and automated manufacturing.

These industries will create new production centres, but the resulting GST may still flow disproportionately towards consumption markets, corporate centres and logistics gateways. The gap between the place of production and the place of taxation could widen further.

The central lesson is simple but important. GST tells us where formal taxable transactions are concentrated. Industrial production tells us where productive capability exists. GDP tells us where economic value is generated. None of these indicators can independently describe the full economic strength of a state.

Haryana demonstrates this new economic reality clearly. Its share in national GST is much higher than its share in GDP and industrial output. This reflects a genuine strength in formalisation, income, consumption, logistics and corporate activity. But it should not be mistaken for complete industrial transformation.

The real test for Haryana is whether it can convert its exceptional tax and transaction base into deeper manufacturing, stronger MSME clusters, wider regional development and more productive employment. High GST collection is an achievement. Turning that fiscal strength into broad industrial capability will be the much larger achievement.

#GST #Haryana #IndustrialDevelopment #Manufacturing #MSME #StateEconomy #EconomicPolicy #ClusterDevelopment #IndiaEconomy