Policy update
Varalika Raizada
INTRODUCTION
Artificial Intelligence has become the game changer that has the potential to transform economies, improve governance and help address critical societal issues. AI is both a technological upgrade and strategic step towards inclusive growth, effective public service delivery and enhancing competitiveness at the global level. AI- based predictive analytics is one such revolutionary technology that utilises data, algorithms and machine learning models to provide predictions of outcomes, forecast trends and support proactive decision-making.
From controlling crowd flows in cities, forecasting traffic flow patterns, or detecting disease breakout hotspots, AI- based predictive analytics is becoming a critical piece of technology in India’s AI push. To leverage its potential India has been actively developing policies and initiatives to utilise the potential of AI across industries like healthcare, agriculture, education, urban governance and transportation. With a strategic vision based on innovation, digital empowerment, and ethical deployment of AI, India’s changing strategy will ensure feasibility and effective use of technology.
INDIA’S EVOLVING APPROACH TO ARTIFICIAL INTELLIGENCE
In India, Artificial Intelligence is increasingly being acknowledged as a force behind social and economic progress. India aims to utilize AI’s potential for responsible innovation and inclusive progress propelled by the #AIforAll vision. According to NITI Aayog’s 2018 national strategy for Artificial Intelligence, AI can help boost India’s international competitiveness . A report by Accenture estimates that by 2035, AI may boost India’s GDP by approximately 957 billion USD, confirming its strategic relevance. In an effort to solve local issues, India has made the capturing of AI technology a national priority.
With the initiation of the IndiaAI Mission in 2024, with the budget of ₹10,371.92 crore across five years, is a significant move to strengthen India’s AI capabilities. Its goal is to ensure adoption of safe and condensed AI practices while developing indigenous prototypes and a public AI infrastructure. In order to boost research and development in AI technology, the government has established the International Centres for Transformational AI (ICTAI) and Centres of Excellence for AI.
Dialogue among relevant stakeholders including tech industry, governments, AI developers, researchers and ethicists is also crucial for optimal and optimal usage of AI in governance. The Responsible AI for Social Empowerment (RAISE) summit is one such initiative under the Ministry of Electronics and Information Technology (MeitY). The national AI mission further aids in providing digital infrastructure, data ecosystem and public private partnership in AI technology.
India’s strategy is to gain competence with global trends; for instance, the UAE has a ministry of AI, and the United Kingdom has an office for AI and an AI council. China, France, US and Japan are significantly boosting public investments in AI R&D startups and infrastructure. India’s approach aims to become a global leader in AI.
The Stanford AI Index 2024 ranks India first globally in AI skill penetration bypassing the United States. By 2035, the AI sector in India is expected to grow to a value of 28.8 billion dollars, potentially contributing $1 trillion to the country’s GDP, according to a report by NITI Aayog. The need for AI specialists is also expected to increase concurrently, reaching one million by 2026.
UNDERSTANDING AI PREDICTIVE ANALYSIS AND ITS ADVANTAGES
Predictive analytics is a potent area of Artificial Intelligence that forecasts trends and makes predictions about the future using statistical models, machine learning algorithms, and historical data. Over time, the models are improved to increase the accuracy of their predictions. The result is a system capable of forecasting the outcomes with remarkable precision. AI-based visitor analytics is increasingly being adopted across various public environments because of its ability to forecast crowd dynamics. Applications include peak travel hours at railway stations, metro terminals, and airports, predictions of crowds at Tourist and Heritage Sites, Malls, and Marketplaces and Religious gatherings and festivals for better safety protocols and resource deployment.
Advantages of AI- based Predictive Analytics
- Video feeds and sensor data are analyzed to monitor crowd density and movement in real time.
- Anticipating peak times and potential congestion points enables authorities to take preemptive action.
- AI models simulate various scenarios based on the predicted crowd behavior, and aid in emergency preparedness.
- AI aids in the smart allocation of resources from sanitation services to law enforcement and traffic management.
- AI models process large datasets quickly, revealing trends and anomalies that may be missed by traditional methods.
- Predictive analytics streamlines operations and minimizes waste.
- Predictive maintenance lowers repair costs and downtime by seeing possible equipment breakdowns before they happen.
APPLICATION OF AI- BASED PREDICTIVE ANALYTICS OF VISITORS TO PUBLIC PLACES AND PEOPLE MANAGEMENT
India has already witnessed successful implementation of pilot and full-scale rollots of AI-powered crowd management systems. These cases highlight the feasibility and effectiveness of the technology.
- Kumbh Mela 2025
The religious event of Mahakumbh in 2025 was the largest human gathering in the world. Determined to tackle the enormous inflow of human crowd and keeping in mind the complexity of the event, authorities resorted to AI-enabled surveillance and analytics, revolutionizing crowd management in India. At the core of the operations was the Integrated Command and Control Centre (ICCC) in Prayagraj, which analyses real-time information from 500 AI-enabled cameras and drones. These systems were designed to monitor video feeds to graph crowd density, monitor human flow across entry and exit points and trigger alerts when overcrowding creates safety issues.
The two main AI models being used were-
- Crowd density mapping: monitors headcount per frame throughput zones and collates information from multiple feeds
- People counting cameras: installed at primary entrances and exits to account for people crossing specific areas.
Although effective, the system is not without flaws. The precision now at 90-95% can be affected due to environmental factors, camera placements and multiple visits by the same person leading to double counting. Despite this, Kumbh Mela offers a rare chance to train AI-models in real-world large-scale human activity. As technology advances, its application to providing safety and efficiency at giganti public gatherings such as Kumbh mela will only become more important. Similarly, recently Lucknow Police used special AI Drones to monitor the Eid celebration at Mosques as security measures.
2. Indian Railways AI integration
Indian railways have started implementing AI-driven passenger flow forecasting models at key high-traffic stations like New Delhi, Howrah, and Mumbai CST to improve crowd management and passenger safety. These technologies are especially beneficial during train delays, when passenger congestion is at peak. A war room at the centre supervises 35 stations to monitor response and coordination in real time.
At Mumbai CST rollout of real time crowd analytics in 2022 resulted in 23% shorter platform clearance times and 15% lowering of crowding complaints. To further enhance infrastructure, Indian railways intends to build 60 permanent holding areas along with AI powered surveillance and predictive analytics to assist in the management of massive crowds and averting crises. This comes after a horrific stampede at the New Delhi railway station as result of an unexpected suge of pilgrims during the Maha Kumbh festival.
- Smart Cities Mission
Numerous smart cities like Surat, Visakhapatnam, and Varanasi are increasingly using AI in urban governance to improve overall functioning of the city, and improve service delivery to the citizens. Under Smart Cities Mission, it is being primarily used to support e- governance, waste management and ensuring public safety.
AI powered smart parks help with park maintenance, optimizing light usage and streamlining other operations, resulting in cost savings. AI is also being used for predicting service delivery by analysing migration trends. In Visakhapatnam, AI is being used to monitor overcrowding and detect unattended baggage through sensor systems that trigger alerts. In Varanasi, AI tools are being used to identify littering and check the status of waste bins for upkeep of public spaces. During the COVID-19 pandemic, AI was implemented in the New Kolkata CB market to detect people not wearing masks and issue them automated warnings.
- Urban Mobility
Urban population in India face constant issues relating to congestion, accidents, underdeveloped public transport and environmental problems. High congestion and improper public transport facilities results in economic inefficiency. In 2015, a staggering 5,00,000 road accidents were reported, claiming 1,46,000 lives. AI based applications in public space and mobility management are becoming increasingly important in this issue.
Intelligent Transportation Systems (ITS), driven by AI are being used to improve traffic flow. In Surat, Public transport vehicle planning and scheduling system for the city bus service uses AI to study real time demand of various routes and studying passenger boarding and exiting. In Visakhapatnam and Nagpur, Intelligent Traffic Management Systems (ITMS) and Video Analytics (VA) are used to predict real time traffic conditions, easing traffic congestions and identifying signal violators.
CHALLENGES
AI based solutions hold great benefits for crowd control and people management. However there are a number of challenges inhibiting its mass implementation.
- Data and Privacy Issues: AI technology poses fear among the public of being continuously watched, creating privacy issues. Having unambiguous privacy and ethical policies make it even more difficult for public acceptance.
- Limited infrastructure and higher implementation expenses: majority of the smaller cities are plagued with lack of modern surveillance systems, weak internet connectivity and insufficient data ecosystems, preventing proper implementation. The costly nature of AI technology along with lack of awareness among common man leads to reluctance to invest.
- Weak innovation methods and shortage of skilled workers: there is a huge shortage of skilled workers with specialised knowledge of AI and data science. The low investments in AI research and challenges in scaling up innovations are still ongoing
- Lack of high quality data: There is also a lack of well structured and high quality datasets, which are required for training and efficacy of AI models, resulting in possible algorithmic bias and decreased accuracy.
- Lack of Stakeholder involvement: failure to bring together the relevant stakeholders including government, industry and academia may decelerate efforts. Unsavoury intellectual environments in the form of unattractive intellectual property regimes further discourage research and innovation in AI.
Overcoming these challenges involves purposeful involvement of the government funding, public-private partnership, focused capacity development and formulation of well articulated regulations regimes to catalyze a healthy and responsive AI ecosystem.
WAY FORWARD
AI- based predictive analytics, especially for controlling public places and population movement, has immense potential. In order to unlock this vision, India needs to take a strategic, inclusive and institutionalized initiative.
- Strengthening AI research ecosystem:
India needs to invest more in AI research, developing context-specific solutions such as dynamic crowd monitoring or mobility smart systems. Cross-sector collaborations and innovation challenges can inspire real-world application and testing, speeding up development.
- Facilitating data access and infrastructure:
The efficient functioning of AI depends heavily on clean and structured data. The government has already launched the IndiaAI dataset platform to ensure the provision of open data repositories across different domains. Accessibility to this AI data platform needs to be ensured specially for research institutions to experiment and scale predictive solutions.
- Upskilling India’s workforce for AI:
India’s youth could play a role as a catalyst for change. Embedding AI and data literacy in public administration, urban planning, and engineering education is imperative. At the same time, targeted reskilling initiatives among current public sector workers and vocational AI training in tier 2 and tier 3 cities can provide inclusive access to an AI-led future.
- Creating responsible and ethical AI
Creating responsible and ethical AI requires developing a custom-designed privacy policy. Establishment of municipal and national AI ethics boards can assist in assessing use cases entailing surveillance or singular data. Open-source, auditable AI solutions for crucial missions like crowd detection, stampede prevention, and emergency alerts will ensure public confidence and support.
India has enormous potential to lead the way in applying AI for the good of society. India may serve as a global model for the responsible application of AI in a vast and diverse nation by making deliberate investments in research, data infrastructure, upskilling employees, ethical control, and institutionalization.
REFERENCES
- Accenture (2021, April 26). REWIRE FOR SUCCESS: BOOSTING India’S AIQ ENABLING STRONG AND INCLUSIVE AI-DRIVEN ECONOMIC GROWTH. https://www.accenture.com/content/dam/accenture/final/a-com-migration/r3-3/pdf/pdf-153/accenture-ai-for-economic-growth-India.pdf
- Banerjee, A. (2025, February 17). Use of AI, permanent holding zones: How railways plans crowd control days after stampede- News18. News18. https://www.news18.com/India/new-delhi-railway-station-stampede-Indian-railways-ai-permanent-holding-areas-ticket-sales-latest-news-9230139.html
- Chakrabarty, S. (2025, January 20). Maha Kumbh 2025: Authorities employ AI to handle rush, predict crowd surges. The Hindu. https://www.thehindu.com/news/national/uttar-pradesh/maha-kumbh-2025-authorities-employ-ai-to-handle-rush-predict-crowd-surges/article69120210.ece
- Desk, T. C. (2025, March 31). Watch: Lucknow Police uses special AI drones to monitor crowd celebrating Eid-ul-Fitr at mosques. The Times of India. https://timesofIndia.Indiatimes.com/city/lucknow/watch-lucknow-police-uses-special-ai-drones-to-monitor-crowd-celebrating-eid-ul-fitr-at-mosques/articleshow/119793481.cms
- How AI-based Traffic Solutions are Improving Traffic Conditions, Reducing Congestion, and Enhancing Public Safety in Cities. (2024, November 6). https://www.itvoice.in/how-ai-based-traffic-solutions-are-improving-traffic-conditions-reducing-congestion-and-enhancing-public-safety-in-cities
- Kumar, A. (2025, February 25). AI meets Maha Kumbh: Demystifying crowd-counting process at Prayagraj. India Today. https://www.Indiatoday.in/India/story/artifical-Intelligence-demystifying-crowd-counting-process-prayagraj-2685514-2025-02-25
- Ministry of Housing and Urban Affairs & Government of India. (2021). Artificial Intelligence USE CASE COMPENDIUM. In MOBILITY (pp. 1–3). https://www.nitiforstates.gov.in/public-assets/Best_Practices/Compendiums/Artificial%20Intelligence%20use%20case%20compendium.pdf
- NITI Aayog (2018, June). National Strategy for Artificial Intelligence. https://www.niti.gov.in/sites/default/files/2023-03/National-Strategy-for-Artificial-Intelligence.pdf
- Press Information Bureau. (2025, March 06). India’s AI Revolution: A Roadmap to Viksit Bharat. Government of India. https://pib.gov.in/PressReleaseIframePage.aspx?PRID=2108810#:~:text=India%20Ranks%201st%20in%20Global,as%20a%20major%20AI%20hub.
- Smart Cities Mission. IndiaAI. https://Indiaai.gov.in/missions/smart-cities-mission
About the contributor– Varalika Raizada is a Research intern at IMPRI.
Acknowledgement– The author extends her sincere gratitude to Aasthaba Jadeja and fellow interns, who provided guidance throughout the process.
Disclaimer: All views expressed in the article belong solely to the author and not necessarily to the organisation.
Read more at IMPRI:
National Supercomputing Mission-2015
National Urban Livelihoods Mission 2013: Creating a safe haven for the urban poor
