# Services, Urbanisation and AI for UPSC: Services Share, GCCs, Smart Cities, Urban Finance and India's AI Strategy (Survey 2025-26) Source: https://tracknation.in/story/services-urbanisation-and-ai-for-upsc-services-share-gccs-smart-cities-urban-finance-and-india-s-ai-strategy-survey-2025-26 (TrackNation β€” free UPSC notes with PYQs) > Services, cities and AI for UPSC β€” services at 53.6% of GDP, the 7th-largest services exporter, GCCs and data centres, the Survey's bottom-up AI strategy and data framework, the Paris and New Delhi AI summits, how cities are defined, governed and financed (UIDF, Urban Challenge Fund), Smart Cities, metro and waste numbers β€” with practice questions, real PYQs and exact Mains questions. ## Services: The Engine of Growth, Exports and FDI 🎯 **Exam priority: Tier-3 as a topic, but the questions are arriving through AI, space, ports, e-commerce and city governance.** *'The services sector' itself is rarely a Prelims topic, yet recent papers keep asking its parts: the 2020 paper asked what AI can effectively do, 2024 asked where municipal powers sit in the Constitution (Part IX-A), 2025 the Paris AI Action Summit and 2026 the New Delhi AI Impact Summit, the IN-SPACe agency, Vizhinjam port and the dropshipping model. Mains has repeatedly asked about Smart Cities, urban local bodies, urban floods and solid waste, and in 2026 'agentic AI'. Learn the Survey's numbers and its argument.* The Economic Survey 2025-26 gives services three chapters: services (7), the AI ecosystem (14) and urbanisation (15). They fit together β€” services are the engine, AI is the technology that may rewrite how the engine works, and cities are where most of it happens. ## Services: the principal engine of growth - **A service-led path at a low income.** Most countries industrialise first. India has instead experienced **service-led growth at a significantly lower level of per capita income** than manufacturing-led peers. Services were **49.9% of GDP in 2024** on World Bank data β€” about 1.5 percentage points above the pre-pandemic average, a bigger rise than the world or the advanced economies β€” and their share in H1 FY26 reached **53.6%**, with services GVA growing **9.3%** against 7.0% a year earlier. - **High growth, low volatility.** Services have grown at an average of about **7-8% a year**. The standard deviation of growth over FY05-FY25 is just **1.22** for services against 4.20 for manufacturing, 3.07 for industry and 1.5 for overall GDP β€” which is why the Survey calls services a high-growth, low-volatility anchor. - **Trade and investment.** India is the **world's seventh-largest exporter of services**; its share of global services trade has more than doubled from **2% (2005) to 4.3% (2024)**. Services remain the **largest recipient of FDI**: on UNCTAD's classification of announced greenfield projects they took **80.2%** of total FDI in FY23-FY25 (77.7% before the pandemic), led by information and communication (25.8%) and professional services (23.8%), then finance and insurance (14.2%), energy and gas (12.8%) and trading (12.2%) β€” nearly 89% together. ## Services exports: software, consulting and the AI effect - Services exports averaged **9.7% of GDP in FY23-FY25** (7.4% before the pandemic) and their growth more than doubled from **7.6% to 14.0%**, before moderating to **8.0% in April-November FY26** amid global competition and protectionism. They act as a buffer when goods trade is weak. - **Software services** are over 40% of services exports and grew 13.5% a year in FY23-FY25 (4.7% earlier). **Professional and management consulting** is the second driver β€” growth of 25.9% lifted its share from 10.5% to 18.3%. Together they are more than 65% of services exports. - **AI and exports.** A difference-in-differences exercise in the Survey finds that AI-intensive services (software, business and financial services) grew about **39.5%** faster than less AI-exposed ones (travel, transport, insurance and others) after the AI diffusion phase β€” about 44% for software and 67% for business services, while the effect for financial services was positive but not statistically significant. The WTO estimates AI could raise global trade by 34-37% by 2040, with digitally deliverable services up about 42%. - **Trade agreements as services policy.** The India-UK CETA (concluded July 2025) opens 137 services sub-sectors, with mutual-recognition agreements for nursing, accountancy and architecture and a Double Contribution Convention that ends dual social-security payments for postings up to 36 months. The India-Oman CEPA (December 2025) raises the intra-corporate transferee quota from 20% to 50%, and the India-EFTA TEPA has been in force since October 2025. The jargon to know: **Mode 1** (cross-border delivery), **Mode 3** (commercial presence) and **Mode 4** (movement of natural persons). - **The Survey's caution.** Services exports are 'firm-selective' rather than 'system-forcing': a software firm can thrive without good ports or courts, so, the Survey argues, services exports cannot by themselves force the state to upgrade its capacity the way goods exports do. It also notes **servicification** β€” domestic services value added is about 17.7% of India's manufacturing export value (2020) β€” so manufacturing and services are complements. ## Sub-Sectors, Jobs and Regions: Tourism, IT, Telecom, Transport, Housing, Media and Space Services is a heading over very different activities. The Survey takes seven sub-sectors; each has one or two numbers worth remembering. ## Jobs, regions and the 'employment-output paradox' - NITI Aayog's October 2025 study broke services into **15 sub-sectors**. Services give around **30% of employment** β€” more than manufacturing and construction (about 11-12% each) β€” and their employment elasticity over 2011-2024 was **0.43**, against 0.22 in manufacturing, 0.41 in agriculture and 0.60 in construction. About **40 million jobs** were added in six years (elasticity 0.63 post-COVID). Yet high-end services produce output faster than jobs, so most new jobs are in low-value trade, hospitality, transport and personal services β€” the **employment-output paradox**. - Globally the services share of employment rose from 35.5% (1992) to 49.8% (2022); India's rose from 22.1% to 30.0%. About 51.5% of service workers are in regular wage jobs, so job quality is better than in agriculture. - Computer and information services doubled their GVA share from 6.0% to 12.2%. Karnataka, Maharashtra, Tamil Nadu and Telangana produce nearly 40% of services output; Kerala (64.3%) and Bihar (58.7%) both have high services shares but of very different quality. In urban India, services employ **61.9%** of workers (PLFS, H1 FY26), and services were 51.7% of net EPFO additions in April-July FY26. - High-frequency signals: the **PMI Services** index averaged 58.9 in Q3 FY26 (above 50 means expansion) and bank lending to services grew 12.4% in October-November 2025. ## The sub-sectors in numbers | **Sub-sector** | **What to remember** | | Tourism | 5.22% of GDP and 8.46 crore jobs (13.3% of employment) in FY24; foreign exchange earnings USD 35.0 billion in 2024 (+8.8%); domestic visits up 17.5% in 2024; international arrivals 20.57 million (+8.9%). Medical-tourism arrivals grew at about 12.4% a year since 2009. Schemes: Swadesh Darshan 2.0, Challenge-Based Destination Development, PRASHAD, UDAN | | IT-ITeS and GCCs | Revenue USD 283 billion in FY25 (+5.1%). India hosts over 1,700 Global Capability Centres β€” offshore units of multinationals β€” employing over 19 lakh people, the largest captive-centre hub in the world. About 32,000-35,000 start-ups (third-largest ecosystem); GenAI start-ups rose from about 240 to over 890 in a year; Tier-1 on ITU's Global Cybersecurity Index (98.49); UNCTAD Frontier Technology Readiness rank 48th to 36th of 170 | | Data centres | Capacity about 1.4 GW (Q2 2025), projected 8 GW by 2030. India generates nearly 20% of world data but hosts only about 3% of data centres (150 of 11,000); energy and water are the constraints | | Transport | About 4.5% of GVA. Port cargo 1,052 to 1,603 million tonnes (FY15-FY25); container-vessel turnaround 43 to 30 hours; air passengers 411.8 million in FY25 (+9.4%); rail is about 50% more cost-effective than road for bulk freight | | Telecom | About 1.2% of GVA; connections 933 million (2014) to over 1.2 billion; tele-density 75% to 86.8%; internet subscriptions 25 crore to 101.8 crore; data price about β‚Ή300 to β‚Ή8.3 per GB | | Real estate and housing | About 7% of GVA over a decade; individual housing loans more than tripled from β‚Ή10 lakh crore (2015) to over β‚Ή37 lakh crore, raising housing credit from 8% to over 11% of GDP. RERA, GST, PMAY-U and the Affordable Housing Fund drove formalisation | | Media and entertainment | About β‚Ή2.5 trillion in 2024, with digital media about one-third. The 'Orange Economy' β€” creativity, culture and intellectual property; live concerts need 10-15 clearances, so a single-window mechanism is being prepared | | Space | USD 8.4 billion (about 2% of the global market) heading to USD 44 billion; 393 foreign satellites launched for 34 countries (2015-2024); NSIL revenue β‚Ή322 crore (FY20) to β‚Ή2,940 crore (FY23); Indian Space Policy 2023 and IN-SPACe opened the sector to private firms; NavIC is India's navigation system | ## Artificial Intelligence: India's Strategy, Governance and Data Framework Chapter 14 is the Survey's argument for how India should approach AI. It is unusually policy-heavy, which makes it good Mains material and a source of Statement-I/II questions. ## Where AI stands, and the six trade-offs - **Adoption is wide; the frontier is narrow.** McKinsey found **88%** of surveyed organisations using AI in at least one function in 2025 (31% scaling it, 7% fully integrated), and high-income countries account for 58.4% of use. But building frontier models needs capital, compute, data and energy that few firms hold β€” and over 70% of data centres (73%) sit in high-income countries, against 3% in India. On jobs, early US evidence (Yale's Budget Lab) shows no discernible disruption, but the Survey's own US analysis finds employment responding less to output growth after late 2022 β€” 'a quiet, steady drift' in the labour intensity of GDP. - **Six asymmetries** frame India's choices: (1) **frontier versus application**; (2) **scale versus inclusion** β€” AI lifts the productivity of capital over labour in white-collar services; (3) **open versus proprietary** models; (4) **compute intensity versus resource constraints** β€” data centres use electricity and water (up to 20 lakh litres a day each); (5) **regulation versus innovation**; (6) **strategic autonomy versus global integration**. - A stress test of India's compute build-out found that **hardware access (GPUs)** β€” not price or grid power β€” becomes the central limiting factor, even with strong demand and cheap finance. Hence a medium-term reliance on diversified access to global compute while the National Semiconductor Mission builds domestic capability. ## The bottom-up strategy - The Survey rejects chasing frontier-scale models and proposes a **bottom-up** path: **application-specific small models** that run on local hardware, built on **open-weight platforms** under the **IndiaAI Mission**, plus an **'AI-OS'** public-good layer in which the sovereign is a shareholder β€” like UPI and Aadhaar β€” pooling data, shared compute and a government-hosted code repository. The ambition is to turn India from the world's 'IT back office' into one of its 'AI front offices'. - **Frugal AI already exists**: AI-enabled agri networks serving 1.8 million farmers in 12 States, low-cost cancer screening, landslide alerts in the Himalaya, and language tools such as **Bhashini** (MeitY) and **AI4Bharat** (IIT Madras). - **People.** 'Earn-and-learn' fellowships integrating school, vocational and early tertiary years from class 11, supported by NEP 2020's multiple entry-exit and Academic Bank of Credits, and by the **Viksit Bharat Shiksha Adhishthan Bill, 2025**; stronger foundational skills in primary school; and skilling for high-skill physical jobs like nursing and geriatric care. ## Governance: regulate in sequence, not all at once - Global models differ: the **EU AI Act** is an omnibus law, China regulates by application, and the US relies on voluntary principles. The Survey favours a **light, incentive-based, risk-weighted** approach built on MeitY's governance guidelines, with oversight embedded in **existing sectoral regulators** rather than a single omnibus AI law. It proposes an **AI Economic Council** to calibrate the pace of adoption to jobs and skills, and an **AI Safety Institute** for risk analysis, red-teaming and international cooperation. - **Data.** Over 100 crore people have broadband, so India's data is a strategic asset. Building on the **DPDP Act, 2023**, the Survey proposes **accountable portability over rigid localisation**: data may cross borders, but large processors must keep auditable records and a **mirrored copy** of relevant datasets in India; obligations scale with risk and size (start-ups and researchers face lighter rules); and firms extracting value from Indian data contribute through a menu β€” local model training, R&D funding, data or compute sharing. - **Sequence.** Phase one builds *coordination* (shared compute, repositories, data categorisation); phase two *capacity* (certified domestic infrastructure); and only then *binding policy leverage* β€” "coordination first, capacity next, and binding policy leverage last". ## Urbanisation: Definitions, Governance and Urban Finance Chapter 15 treats cities as **economic infrastructure**. Its core claim: India is far more urban in economic terms than its statistics show, but its cities lack the powers, money and plans to match their importance. ## How urban is India? - **Census definition.** A place is urban if it has a population above **5,000**, at least **75% of male workers in non-agricultural work** and a density of at least **400 persons per sq km** β€” plus statutory towns (those with a municipal body). The system is top-heavy: over 70% of urban residents lived in Class I cities (above one lakh) in 2011, and the **52 metropolitan cities** held 42.3%. The RBI classifies cities into six tiers by population (Tier 1: one lakh and above). - **The measurement gap.** Using satellite-based **DEGURBA** methods, India was **63% urban in 2015** β€” nearly double the Census 2011 figure; for Kerala, the Survey finds urbanisation of about 82.6% in 2025 against 47.7% in the 2011 Census. Growth is **outward**: in 16 cities the periphery grew faster than the core between 2000 and 2020 (a periphery-to-core ratio above one). - **Why cities matter.** Doubling a city's size raises productivity by about **12%** in India (agglomeration economies). The World Bank expects 600 million urban residents β€” **40% of the population** β€” by 2036, from 31% in 2011, producing almost 70% of GDP. The top ten cities hold about 9% of the population and 28% of GDP. ## The governance and finance deficit - Urban functions are split among **urban local bodies, development authorities, State line departments and parastatals**; land use, policing and utilities often sit with the State. The 74th Amendment (Part IX-A of the Constitution) created municipalities, but powers remain thin β€” hence the Mains question on States' reluctance to empower urban local bodies (2023). - Indian cities raise **less than 0.6% of GDP** as own-source revenue. Urban property tax is only **0.15% of GDP** against 0.3-0.6% in low- and middle-income countries; own revenue covers just 30-40% of spending in large cities and under 20% in smaller ones, and borrowing is negligible. 'Global cities compete; Indian cities comply.' - **Financing windows.** The **Urban Infrastructure Development Fund** (Budget 2023-24, initial β‚Ή10,000 crore, run through the National Housing Bank) lends to Tier-2 and Tier-3 cities; the **Urban Challenge Fund** (Budget 2025-26, β‚Ή1 lakh crore) is performance-linked β€” it co-finances up to **25%** of a bankable project, provided at least **50%** comes from bonds, bank loans and PPPs. The World Bank puts India's urban infrastructure need at USD 840 billion over 15 years (about USD 55 billion a year). The Survey also suggests a statutory **20-year City Spatial and Economic Plan** for every million-plus city, self-updating property tax and municipal bonds without State guarantees. ## Urban Land, Housing, Mobility, Waste and the Flagship Missions ## Land, housing, mobility and waste: the binding constraints - **Land as 'dead capital'.** Low **floor space index (FSI)** under Development Control Regulations caps built-up area per plot, pushing cities outwards and raising prices; unclear titles block redevelopment. The fixes are digital land records (**DILRMP**, the Unique Land Parcel Identification Number or **ULPIN/Bhu-Aadhaar**, the National Generic Document Registration System) and rule-based, transit-linked density β€” with the warning that density without services only creates gridlock. - **Housing.** Under the two phases of **PMAY-Urban**, 122.06 lakh houses were sanctioned and 96.02 lakh delivered by November 2025. The urban housing deficit was 18.8 million in 2012 and was estimated at 29 million in 2018, almost all in low-income groups. Affordable housing tends to be built on city fringes lacking transit and services. - **Mobility.** Congestion costs are large (Uber-BCG: USD 22 billion a year across four metros; TomTom: 117, 103 and 76 hours lost a year in Bengaluru, Mumbai and New Delhi). The principle is to design for **people, not vehicles**, as the National Urban Transport Policy (2014) says. About **1,036 km of Metro/RRTS** run in around 24 cities under the **Metro Rail Policy (2017)** and the **National TOD Policy (2017)**. The **Namo Bharat RRTS** (Delhi-Meerut, 82 km, up to 180 km/h) is funded 20% by the Centre, 20% by States and 60% by ADB, AIIB and NDB. **PM e-Bus Sewa** backs 10,000 e-buses under PPP with β‚Ή20,000 crore of central assistance and a payment security mechanism; yet MoHUA's norm of 40-60 buses per lakh people is unmet β€” India has only about 47,650 city buses, 61% of them in nine megacities. Singapore's Area Licensing Scheme (1975) and London's congestion charge (2003, traffic down 16%) show demand management works. - **Waste and water.** Swachh Bharat Mission-Urban 2.0 (October 2021) moved from access to outcomes: door-to-door collection reached 98% of urban wards, and about 1.6 lakh tonnes of waste a day is generated with around 80% reported processed (16% in 2014-15). **Garbage-Free City** ratings run from 1 to 7 stars (a 3-star city needs 60% segregation and 70% processing; 7-star needs 90% and near-complete legacy-dump remediation). Of over 16 lakh Cleanliness Target Units, 15.74 lakh have been transformed. **Indore** went from 25th (2016) to first every year since 2017 through behaviour change, self-help groups and enforcement. Of 112 billion litres of wastewater a day, only 28% is treated in cities and 8% of that reused. - **Flagship missions.** The **Smart Cities Mission** (June 2015, 100 cities) has completed over 90% of about 8,067 projects (β‚Ή1.64 lakh crore), and all 100 cities have Integrated Command and Control Centres. **AMRUT** and AMRUT 2.0 cover water and sewerage. MoHUA's **Ease of Living Index** (2025) was topped by newer, Tier-2 cities β€” Pune, Navi Mumbai, Greater Mumbai, Tirupati and Chandigarh among them. ## How UPSC Could Ask This Next ## How UPSC could ask this next - **The summit-detail stem.** 2025 and 2026 asked about the Paris AI Action Summit (co-chaired by India; the US and UK did not sign) and the New Delhi AI Impact Summit (People, Planet, Progress; seven 'Chakras'; a non-binding charter). Expect the next one to turn a single word β€” 'binding', 'mandatory', 'all countries' β€” into the trap. - **The 'which agency' stem.** IN-SPACe authorises private space activity; NSIL is the commercial arm; NavIC is the navigation system; the AI Economic Council is a Survey *proposal* and the AI Safety Institute a function the MeitY governance guidelines propose β€” neither is a statute or a statutory body, so do not mark them as existing laws. - **The near-miss numbers.** 53.6% is services' share of GDP (H1 FY26), 58.9 is the services PMI, 51.7% is the services share of EPFO net additions, 61.9% is services' share of urban employment β€” four different numbers that look alike. Services FDI share (80.2%) differs from the DPIIT figure because UNCTAD uses a broader definition. - **The near-miss institutions.** UIDF (Budget 2023-24, through NHB, β‚Ή10,000 crore) versus Urban Challenge Fund (Budget 2025-26, β‚Ή1 lakh crore, co-finance up to 25%). Smart Cities Mission (100 cities, 2015) versus AMRUT (water and sewerage). Part IX-A (Municipalities) versus Part IX (Panchayats). - **Mains angles.** (1) Services-led growth: boon or premature deindustrialisation? (2) How should India approach AI β€” frontier models or application-led? (3) Why are Indian cities economically central but politically weak? (4) Smart Cities Mission β€” success or failure? (5) Agentic AI: benefits, risks, regulation. ## Further Reading - NCERT Economics textbooks (Class 11–12), the Economic Survey, the Union Budget documents and RBI publications cited in this note β€” the primary sources behind every figure above. ## Test Yourself: Practice Questions and PYQs Twelve practice questions on services, AI and cities, followed by real UPSC questions (some from the 2026 paper, whose official key is provisional) and exact Mains questions. ## Practice Questions **Q1.** According to the Economic Survey 2025-26, the share of services in India's GDP in the first half of FY26 rose to - (a) 49.9 per cent - (b) 47.1 per cent - (c) 58.9 per cent - (d) 53.6 per cent **Answer: (d)** β€” Services were 53.6 per cent of GDP in H1 FY26. 49.9 per cent was the World Bank-based share for 2024 and 58.9 is the PMI Services reading for Q3 FY26 β€” the two near-miss numbers. **Q2.** Consider the following statements: 1. India is the world's seventh-largest exporter of services. 2. Software services account for over 40 per cent of India's services exports. 3. Services have been the largest recipient of foreign direct investment in India. 4. Services growth has been more volatile than industry growth. How many of the statements given above are correct? - (a) Only one - (b) All four - (c) Only three - (d) Only two **Answer: (c)** β€” Statements 1, 2 and 3 are correct. Statement 4 is wrong: the standard deviation of services growth (1.22) is far lower than that of industry (3.07), which is why services are a low-volatility anchor. **Q3.** According to the Economic Survey 2025-26, which of the following figures represents the share of services in urban employment (PLFS, first two quarters of FY26)? - (a) 41.9 per cent - (b) 61.9 per cent - (c) 51.7 per cent - (d) 71.9 per cent **Answer: (b)** β€” 61.9 per cent. The 51.7 per cent figure is the services share of net EPFO payroll additions in April-July FY26. **Q4.** In October 2025 NITI Aayog undertook a first-of-its-kind disaggregation of India's services sector into how many major sub-sectors? - (a) 15 - (b) 10 - (c) 20 - (d) 5 **Answer: (a)** β€” NITI Aayog's two-volume report separated services into 15 major sub-sectors and found that services give about 30 per cent of employment. **Q5.** Consider the following statements: Statement-I: Global Capability Centres are offshore units set up by multinational firms to carry out technology development, engineering, analytics and business operations for their global activities. Statement-II: India hosts over 1,700 Global Capability Centres employing over 19 lakh professionals, making it the largest global hub for such captive centres. Which one of the following is correct in respect of the above statements? - (a) Both Statement-I and Statement-II are correct and Statement-II is the correct explanation for Statement-I - (b) Statement-I is correct but Statement-II is incorrect - (c) Statement-I is incorrect but Statement-II is correct - (d) Both Statement-I and Statement-II are correct and Statement-II is not the correct explanation for Statement-I **Answer: (d)** β€” Both statements match the Survey. Statement II gives India's scale; it does not explain what a GCC is, so it is not the explanation of Statement I. **Q6.** Despite generating nearly 20 per cent of the world's data, India hosts only about what share of global data centres, according to the Economic Survey 2025-26? - (a) 13 per cent - (b) 23 per cent - (c) 3 per cent - (d) 33 per cent **Answer: (c)** β€” About 3 per cent β€” around 150 out of 11,000 data centres worldwide. Data-centre capacity is projected to rise from about 1.4 GW to about 8 GW by 2030. **Q7.** Which of the following best describes the AI strategy that the Economic Survey 2025-26 recommends for India? - (a) Leaving AI to the market with no public role until the technology matures - (b) A top-down push to build frontier-scale foundation models at any cost - (c) A ban on foreign AI services until domestic models are ready - (d) A bottom-up approach built on application-specific small models and open-weight platforms, with shared public infrastructure **Answer: (d)** β€” The Survey calls a bottom-up approach 'strategically necessary': small, sector-specific, open and interoperable models, anchored by the IndiaAI Mission and shared infrastructure, in place of frontier-scale competition. **Q8.** The Survey's difference-in-differences analysis found that exports of AI-intensive services (software, business and financial services) grew by approximately how much faster than those of less AI-exposed services after the AI diffusion phase? - (a) 39.5 per cent - (b) 9.5 per cent - (c) 59.5 per cent - (d) 19.5 per cent **Answer: (a)** β€” About 39.5 per cent. Within it, software (about 44 per cent) and business services (about 67 per cent) were significant; financial services was positive but not statistically significant. **Q9.** Consider the following statements about the data-governance framework proposed in the Economic Survey 2025-26: 1. It requires all processing of Indian personal data to take place within India. 2. It requires large processors to keep a mirrored copy of relevant datasets in India so that regulators can oversee them. 3. Its obligations scale with the risk and size of the data use, with a lighter regime for start-ups and research institutions. Which of the statements given above are correct? - (a) 1 and 2 only - (b) 2 and 3 only - (c) 1 and 3 only - (d) 1, 2 and 3 **Answer: (b)** β€” Statement 1 is wrong: the framework prefers 'accountable portability over rigid localisation' and proposes no mandate for domestic processing. Statements 2 and 3 are correct. **Q10.** Which one of the following is NOT one of the conditions that the Census of India uses to classify a settlement as urban? - (a) A population density of at least 400 persons per square kilometre - (b) A population greater than 5,000 - (c) The presence of a municipal corporation - (d) At least 75 per cent of male employment in non-agricultural activity **Answer: (c)** β€” The census towns need the three demographic tests. Places with a municipal body are separately counted as statutory towns β€” that is an administrative route, not one of the three conditions. **Q11.** According to the Economic Survey 2025-26, Indian cities raise less than what share of GDP as own-source revenue? - (a) 2 per cent - (b) 4 per cent - (c) 0.6 per cent - (d) 6 per cent **Answer: (c)** β€” Less than 0.6 per cent of GDP; urban property tax is just 0.15 per cent. The 2-4 per cent range is the figure for OECD cities, a near-miss trap. **Q12.** Consider the following statements: 1. The Urban Infrastructure Development Fund was announced in the Union Budget 2023-24 with an initial outlay of β‚Ή10,000 crore and is operated through the National Housing Bank. 2. The Urban Challenge Fund co-finances up to 50 per cent of the cost of a bankable urban project. 3. The Urban Challenge Fund requires at least 50 per cent of project cost to come from bonds, bank loans and PPPs. How many of the statements given above are correct? - (a) Only one - (b) None - (c) Only two - (d) All three **Answer: (c)** β€” Statements 1 and 3 are correct. The Urban Challenge Fund (β‚Ή1 lakh crore, Budget 2025-26) co-finances up to 25 per cent, not 50 per cent, so statement 2 is wrong. ### UPSC Previous Year Questions (PYQs) *Real UPSC Prelims questions, checked against the official paper and answer key. 2026 items use the provisional key and are marked in the explanation.* **Q1.** (2020) With the present state of development, Artificial Intelligence can effectively do which of the following? 1. Bring down electricity consumption in industrial units 2. Create meaningful short stories and songs 3. Disease diagnosis 4. Text-to-Speech Conversion 5. Wireless transmission of electrical energy Select the correct answer using the code given below: - (a) 1, 2, 3 and 5 only - (b) 2, 4 and 5 only - (c) 1, 2, 3, 4 and 5 - (d) 1, 3 and 4 only **Answer: (d)** β€” AI today can help cut industrial electricity use through optimised control (1), support disease diagnosis from images and records (3) and convert text to speech (4). It cannot produce genuinely meaningful short stories or songs on its own (2 is outside what the paper treated as effective), and wireless transmission of electrical energy is a physics and engineering problem, not an AI task (5). So 1, 3 and 4 only. **Q2.** (2025) Consider the following statements regarding AI Action Summit held in Grand Palais, Paris in February 2025: I. Co-chaired with India, the event builds on the advances made at the Bletchley Park Summit held in 2023 and the Seoul Summit held in 2024. II. Along with other countries, US and UK also signed the declaration on inclusive and sustainable AI. Which of the statements given above is/are correct? - (a) I only - (b) II only - (c) Neither I nor II - (d) Both I and II **Answer: (a)** β€” The Paris AI Action Summit (February 2025) was co-chaired by France and India and followed the Bletchley Park (2023) and Seoul (2024) summits, so statement I is right. The United States and the United Kingdom declined to sign the declaration on inclusive and sustainable AI, so statement II is wrong. **Q3.** (2026) Consider the following statements with respect to the AI Impact Summit, 2026 held in New Delhi: 1. The Summit's intellectual framework was based on three foundational Sutras: People, Planning, and Progress. 2. The Preamble of the Summit stresses Democratising AI Resources, which acknowledges the Charter for Democratic Diffusion of AI as a binding framework to support locally relevant innovation and strengthen resilient AI ecosystems while respecting national laws. 3. The New Delhi Declaration on AI Impact was structured around seven Chakras (Pillars), which included Access for Social Empowerment, AI for Science, and Secure and Trusted AI. Which of the statements given above is/are correct? - (a) 3 only - (b) 1 and 2 only - (c) 1, 2 and 3 - (d) 2 and 3 only **Answer: (a)** β€” The Summit's themes were People, Planet and Progress β€” 'Planning' in statement 1 is the planted error. The Charter for Democratic Diffusion of AI is a voluntary, non-binding framework, so statement 2 is wrong when it calls it binding. Statement 3 is correct: the New Delhi Declaration was structured around seven Chakras, including Access for Social Empowerment, AI for Science and Secure and Trusted AI. (Based on the provisional 2026 key.) **Q4.** (2026) An e-commerce revenue model where the seller has control over pricing but doesn't keep products in stock and instead transfers customer orders and shipment details to a third-party supplier, who then ships the goods directly to the customer, is called: - (a) Affiliate Revenue Model - (b) Transaction Fee Revenue Model - (c) Dropshipping Model - (d) Agency Revenue Model **Answer: (c)** β€” In dropshipping the seller sets the price but never holds stock: the order and shipping details pass to a third-party supplier who ships straight to the customer. Affiliate models earn a commission on referred sales, transaction-fee models charge per transaction, and agency models earn a fee for connecting buyer and seller. (Based on the provisional 2026 key.) **Q5.** (2026) Consider the following statements with regard to involvement of private entities in India's space programme: 1. The Indian National Space Promotion and Authorisation Centre (IN-SPACe) is an autonomous agency formed to facilitate participation of private entities. 2. Agnikul Cosmos launched the world's first flight using 3D-printed rocket engine. 3. Skyroot Aerospace has developed liquid fuel for GSLV. Which of the statements given above is/are correct? - (a) 2 and 3 only - (b) 1, 2 and 3 - (c) 1 only - (d) 1 and 2 only **Answer: (d)** β€” IN-SPACe is an autonomous agency that authorises and promotes private participation in India's space activity (statement 1 is right). Agnikul Cosmos flew a rocket powered by a single-piece 3D-printed engine (statement 2 is right). Skyroot Aerospace is known for its Vikram launch vehicles and liquid engines for its own rockets β€” it did not develop liquid fuel for ISRO's GSLV (statement 3 is wrong). (Based on the provisional 2026 key.) **Q6.** (2026) In what way(s) does the Vizhinjam International Seaport represent a structural shift in India's maritime trade and logistics policy? 1. By functioning exclusively as a domestic cargo hub to reduce reliance on coastal shipping and eliminate the need for foreign collaborations. 2. By focusing primarily on passenger cruise tourism and heritage shipping to increase Kerala's profile as a maritime heritage destination. 3. By leveraging its natural deep draft and strategic location to reduce dependence on foreign trans-shipment ports, enhance revenue retention, and reposition India in regional maritime trade. Select the answer using the code given below: - (a) 2 and 3 - (b) 3 only - (c) 1 and 2 - (d) 1 only **Answer: (b)** β€” Vizhinjam is a deep-water transhipment port: its natural draft and location on the main east-west shipping route let India keep transhipment cargo and revenue that used to go to foreign hubs such as Colombo, Singapore and Salalah β€” statement 3. It is not an exclusively domestic hub (1) and not a cruise or heritage port (2). (Based on the provisional 2026 key.) **Q7.** (2024) Which of the following statements are correct about the Constitution of India? 1. Powers of the Municipalities are given in Part IX A of the Constitution. 2. Emergency provisions are given in Part XVIII of the Constitution. 3. Provisions related to the amendment of the Constitution are given in Part XX of the Constitution. Select the answer using the code given below: - (a) 1 and 3 only - (b) 1, 2 and 3 - (c) 1 and 2 only - (d) 2 and 3 only **Answer: (b)** β€” Part IX-A (Articles 243P to 243ZG), added by the 74th Amendment, deals with Municipalities; Part XVIII carries the Emergency provisions (Articles 352-360); and Part XX (Article 368) deals with amendment of the Constitution. All three statements are correct. ### Mains Practice Questions Use these to frame full-length answers. You don't have to answer one exactly β€” they show the angles UPSC tests, so let them guide which points you cover. - India has grown through services rather than manufacturing at a low level of per capita income. Is this a strength or a risk? Discuss the employment-output paradox of the services sector. - What are 'Smart Cities'? Examine their relevance for urban development in India and the lessons of the Smart Cities Mission. - 'The states in India seem reluctant to empower urban local bodies both functionally as well as financially.' Comment, with reference to the finances of Indian cities. - Should India chase frontier AI models or build application-specific, open and small models? Evaluate the Economic Survey's bottom-up strategy and the governance framework it proposes. - What is agentic AI? Discuss its advantages, risks and the challenges of governing it for a labour-abundant economy like India. ### UPSC Mains β€” Previous Years’ Questions on this topic (last 10 years) These are the **actual GS Mains questions** UPSC has asked on this theme β€” real proof of how, and when, it is tested. Try to write a full answer to each. - 2026 (GS-III): What is agentic Artificial Intelligence (AI)? Explain its working. Describe its applications with suitable examples. Discuss the advantages, risks and challenges associated with agentic AI systems. - 2026 (GS-III): What are the challenges to solid waste management in India? Discuss the governmental policy framework on solid waste management. Discuss the success/failure cases of Delhi and Indore cities highlighting the salient feature of their solid waste management initiatives. - 2024 (GS-II): Analyse the role of local bodies in providing good governance at local level and bring out the pros and cons merging the rural local bodies with the urban local bodies. - 2024 (GS-III): Flooding in urban areas is an emerging climate-induced disaster. Discuss the causes of this disaster. Mention the features of two such major floods in the last two decades in India. Describe the policies and frameworks in India that aim at tackling such floods. - 2023 (GS-III): Introduce the concept of Artificial Intelligence (AI). How does AI help clinical diagnosis? Do you perceive any threat to privacy of the individual in the use of AI in healthcare? - 2023 (GS-II): "The states in India seem reluctant to empower urban local bodies both functionally as well as financially." Comment. - 2016 (GS-III): What are 'Smart Cities'? Examine their relevance for urban development in India. Will it increase rural-urban differences? Give arguments for 'Smart Villages' in the light of PURA and RURBAN Mission. - 2016 (GS-III): The frequency of urban floods due to high intensity rainfall is increasing over the years. Discussing the reasons for urban floods, highlight the mechanisms for preparedness to reduce the risk during such events.