Authors: Debashrita Mazumder, Angki Lapung, Aradhya Chaturvedi, Lavanya Aditi Puri, M. Rahul, Sree Nanda S, Khushi Sharma

Abstract

Urban India’s public health system is confronting unprecedented strain due to fragmented infrastructure, workforce shortages, and deep-rooted socio-economic disparities, particularly among marginalized communities. Life expectancy among the bottom 30% of urban dwellers is up to 9.1 years lower for men and 6.2 years lower for women compared to their wealthiest peers.  As India’s cities are projected to house 900 million people by 2047, these disparities are set to intensify.

Artificial intelligence (AI) provides scalable solutions for disease surveillance, diagnostics, and the management of chronic diseases. Tools like Qure.ai’s qXR have improved tuberculosis detection rates by 15.8% in Nagpur, directly augmenting radiologist capacity in resource-limited settings. AI has demonstrated its potential to expand diagnostic reach, enhance chronic disease management, and alleviate clinician workloads. However, access to AI-driven healthcare remains inequitable due to the persisting digital divide. Only 46% of Indians own smartphones, with penetration even lower in urban slums, and digital literacy remains a significant barrier, especially for women and low-income groups 

This review-based paper evaluates AI’s dual role as a catalyst for equitable healthcare and a mirror reflecting systemic inequities. By analysing implementation challenges from infrastructural gaps to socio-cultural resistance, it argues for policy frameworks that prioritise transparency, community-driven design, and intersectional equity to ensure urban health systems serve all citizens, not just the digitally privileged. It draws on mixed-method analysis, combining quantitative data from health NGOs and qualitative interviews with stakeholders to assess how AI is being implemented, resisted, or co-opted within existing public health frameworks.

Introduction: 

Urban India’s public health system is under immense strain, grappling with fragmented infrastructure, workforce shortages, and deeply entrenched inequities, especially among marginalized communities. While cities account for 34% of India’s population, health disparities persist: primary health centres (PHCs) serve 36,000 urban residents per facility compared to 9,000 in rural areas, exacerbating overcrowding and service delays (Patel & Mehta, 2024). The urban poor, comprising 30% of city dwellers, face life expectancies 9.1 years lower for men and 6.2 years for women compared to the wealthiest quintile (Dhabliya et al., 2024). These systemic gaps are compounded by rapid urbanisation, with projections suggesting 900 million urban residents by 2047, intensifying pressure on already strained resources.

Artificial intelligence (AI) emerges as a critical tool to bridge these gaps, offering scalable solutions for disease surveillance, diagnosis, and personalised care. For instance, AI-driven TB screening tools like Qure.ai’s qXR demonstrated a 15.8% increase in case detection in Nagpur by augmenting radiologists’ capabilities in resource-limited settings (Vijayan et al., 2023). Similarly, predictive models leveraging climate data and machine learning enable early warnings for dengue outbreaks, empowering authorities to allocate resources proactively (Yacob & Koll, 2025). Chronic disease management also benefits: AI algorithms personalise treatment plans for diabetes and hypertension, reducing clinician burnout and improving adherence (Singareddy et al., 2023). The Indian government has prioritised AI integration through initiatives like the Ayushman Bharat Digital Mission (ABDM), which establishes a unified digital health ecosystem to streamline telemedicine and electronic health records (Multiple Authors, 2024).

However, urban health disparities persist: women and low-income groups face systemic barriers in accessing AI tools due to smartphone affordability, digital literacy gaps (61% urban vs. 25% rural), and cultural mistrust (Wang et al., 2022). Mental health chatbots like SnehAI, designed for sexual and reproductive education, highlight both AI’s potential to bypass stigma and its limitations. Fewer than 30% of users were women, reflecting gendered access barriers (Varghese et al., 2024).

Ethical challenges further complicate AI adoption. Algorithmic biases in training datasets risk exacerbating diagnostic inaccuracies for underrepresented populations, while data privacy concerns undermine public trust (Chettri et al., 2025). The 2023 Digital Personal Data Protection Act lays the groundwork for governance, but enforcement remains inconsistent. Despite these hurdles, pilot projects like Agent-Based Modelling for water safety in Ireland demonstrate how explainable AI (XAI) can simulate policy impacts on marginalised communities, offering lessons for equitable design in Indian cities (Asghar et al., 2024).

Literature Review :

The scope of public health is developing rapidly, with artificial intelligence (AI) at its core. It is changing the way we provide healthcare, especially in diverse and densely populated countries like India. This review examines AI’s application in public health, which focuses on important factors such as equal access, service distribution to individuals, and policy obstacles associated with AI. We will identify areas where there is a decrease in our existing knowledge, with emphasis laid on intervals in the literature. 

AI in Public Health: An International and Indian Approach 

Consider AI as a strong new ally in the medical field. Worldwide, it is turning from theory to real, solid applications. The AI ​​is helping workers offer immediate help through complex diagnosis, accelerating the development of new drugs, facilitating highly accurate robot surgery and offering immediate help through intelligent chatbots. (Patel and Mehta, 2024). In a world where data and services are only one tap away, the importance of AI is increasing. However, it is important to ensure that AI is used morally. 

India faces specific public health issues, with a population of 1.428 billion individuals distributed in a large area. Our healthcare systems include government programs, private medical facilities and a complicated mix of many NGOs, which are all working to meet the needs of a highly diverse population. However, major obstacles remain: healthcare is lacking in workers, rural infrastructure often lacks adequate, and public knowledge and treatment of diseases can be minimal. This is where AI intervenes, providing a strong array of resources to resolve these systemic challenges and create significant changes. 

A major depiction of AI’s capacity in India is its application in analysing the chest X-ray (CXR) to detect tuberculosis (TB). TB remains an important public health challenge in the country. In 2023, India bore the largest global load, with 26% (Global Tuberculosis Report 2024) in all cases. The National TB Elimination Program (NTEP) recommends the use of CXRS with sputum tests for diagnosis. Nevertheless, this enhancement has proved challenging due to inefficient radiologists and a lack of constrained resources, especially in distant areas. To address the issue, an Indian AI firm, Qure.ai, created QXR. This software, enhanced by AI, helps in analysing CXR images, providing health professionals with important details about TB signals. Research indicates that QXR shows remarkable performance, maintaining sensitivity compared to human radiologists (Vijayan et al., 2023). 

A real success story came out in Nagpur, Maharashtra, under the TB Reich Initiative. This research focuses on the influence of the COVID-19 epidemic on TB information, in combination with India’s ongoing TB issue. The implementation of  QXR in eight private CXR laboratories, using local networks and cloud connectivity, assisted radiologists in identifying potential TB patients. By investigating 10,481 TB cases, AI independently detected an impressive 15.8% increase in TB cases, which was first ignored by human reviewers (Vijayan et al., 2023). This result strongly displays AI’s ability to increase clinical accuracy and case identification in India’s public health system. 

Imagine the AI ​​model that considers climatic factors to predict disease outbreaks like dengue, providing public health authorities an essential opportunity to take action (Yakub and Coal, 2025). A systematic review by Singarddy et al. (2023) emphasised the increasing role of AI in the management of diseases like diabetes, high blood pressure and heart diseases. This is particularly important in urban India, where lifestyle-related diseases are regrettably increasing.

Equity, Accessibility and Policy Enigma 

The actual strength of AI depends on the effectiveness and equity of its integration. It depends on addressing concerns about access, fairness and establishment of a strong policy framework. India’s diverse socio-economic scenario, digital literacy, availability of smartphones and different degrees of cultural practices are characterised by different challenges for comprehensive AI integration. 

Digital division and access barriers:

Even though digital equipment is widely accessible, there is a persistent digital divide, especially in urban India. The risk population, especially women, often faced obstacles such as a lack of education, restricted access to smartphones, and digital literacy (Manvi S et al., 2025; Bhat et al., 2020). Research on digital health technologies in India, especially in cancer treatment, indicates that women and elderly individuals are less inclined to use digital devices, highlighting the required requirement for cultural and linguistic designs (Venkataraman et al., 2022).

Gender use and trust:

Indian mobile health initiative analysis highlights a remarkable hesitation among women in using digital platforms. It often arises in an overall mistrust (passumarthi, 2024) in concerns and health applications about privacy. It highlights the gender-appropriate outreach approach, strong awareness initiative and the requirement for solid privacy guarantees within the AI ​​system. The initiative is originally aligned with the recommendations of the World Health Organisation (WHO), which promotes reducing prejudice and increasing equity in digital mental health programs. Low engagement of women with SnehAi, a Hinglish chatbot for education on sexual and reproductive health, enhances this gender inequality, indicating a more tailored and reference-inconceivable design. 

Governance and policy environment:

To ensure that AI flourishes morally in the public health system of India, it is necessary to establish an intensive and flexible policy and governance structure. With programs such as the first National Digital Health Mission (NDHM) and Ayushman Bharat Digital Mission (ABDM), India has tried to set up a nationwide digital health structure, facilitating  AI integration into public health infrastructure.

The rapid advancement of AI introduces fresh moral issues, especially about data privacy, informed consent and prejudices in algorithms. Policymakers, technology creators and communities should cooperate closely to develop AI moral guidelines. Concepts such as “Health Data collective” and blockchain laser technologies (Chettri et al., 2025) are coming out as an effective solution to communities to keep more control over their health information, build confidence and ensure AI’s moral training. 

Integration of AI in mental health:

Estimates indicate that 70–92% of people with mental health issues lack care (Chaudhary et al., 2024), making AI a powerful solution to remove these inequalities. AI tools such as chatbots (eg, WYSA) and mobile applications help increase access,  reduce the stigma, and provide scalable mental health solutions (Incoster, Sarda, and Subramanian, 2018; Varghese et al., 2024). The Government of India initiatives, such as the Tele-Manas Program and Clinical Decision Support System (CDSS) from Chandigarh,  provide mental health support to underserved urban communities (Das et al, 2024).

Literature Gaps:

Despite increasing research on AI’s abilities in public health, many important aspects require further investigation, especially in the context of India and the comprehensive public health policy. Including AI in current programs: Initiatives like QXR are promising, yet more research is necessary to check the real challenges and achievements of integrating AI solutions in India’s existing public health systems on a large scale. This includes checking interoperability, educating healthcare professionals and securing sustainable financing.

Although some surveys are available, there is a lack of comprehensive qualitative studies, checking, acceptance and confidence levels of frontline healthcare professionals, policy makers, and various community members about AI in public health within various urban environments in India. What are the true views of people on AI in healthcare? 

Shifting from “What” to “How”:

An important part of existing literature emphasises what AI is capable of (diagnosis, prediction). More studies are required on how to apply it fairly. This action enables strategies that effectively reduce prejudice, promote cultural awareness, and include all individuals, especially in various urban Indian populations, long-lasting, extending, expandable solutions. 

Sustainable Economic Model: The diverse health economy across India has a significant lack of research on the financing, maintaining and expansion of AI solutions in public health, especially to address underserved communities without increasing their out-of-pocket costs. 

Addressing these important deficiencies is important to develop public health policies that are actually supported by evidence. This will enable us to efficiently benefit from the remarkable abilities of AI to promote health results for urban communities, guarantee equity, guarantee equity, to guarantee equity.

 Methodology :

In current scholarly literature regarding AI and public health, the overwhelming trend remains in favour of using AI as a solution to specific problems rather than a one-size-fits-all approach. Since AI is a relatively new technological advancement, concerns about regulation and lawmaking still exist, especially in a field as data-sensitive as public health. Our research uses a mixed-method narrative review approach to investigate our claim (Samsurijan et al., 2022). The quantitative aspect of our study covers data from NGOs and researchers, especially in disease prevalence, healthcare utilisation and demographic variables (Danish and Senjyu, 2023). In the qualitative component, we have followed and studied interviews, focus groups and discussions with stakeholders directly involved in the public health industry, such as healthcare providers, medical associates and industry providers. To give the field of AI equal representation, we have incorporated the experiences of policymakers, community leaders and software engineers. This data followed a semi-structured interview protocol, which allows for flexibility and inclusiveness for the experiences of each participant (Olawade et al., 2023) Using strategic integration of quantitative analysis of large datasets with qualitative analysis of stakeholders, we have tried to curate a well-rounded understanding of the multifaceted impact of AI on public health policy. Data was analysed using statistical techniques such as regression, time series, and spatial analysis to identify correlations between AI’s usage and improvements in public health. The selection criteria for these papers were of the utmost importance: we focused on peer-reviewed papers from the period of 2017-2024. These papers, ideally, were to be focused on low and middle-income countries and health-specific sectors. Various databases were scoured, such as DOI, PubMed, Scopus and Google Scholar. To curate a more efficient search, the keywords used to search for supporting papers were “LMIC”, “healthcare” “, Artificial Intelligence” “, public policy” “, strategic” ” ” and “integration.” As our research reveals, we have emphasized including international examples, especially (but not limited to) Southeast Asia, Brazil, and Europe for comparative insights.

Digital health tools and Artificial Intelligence in countering COVID-19 – A global case study 

Covid 19 pandemic has transformed the way we look at Artificial intelligence and digital technology as tools that not only facilitate automation but are a noble innovation for public health infrastructure. As the COVID-19 cases compounded, traditional healthcare scrambled to accommodate the massive influx of patients. The emergency responses and lockdowns put in place by Government authorities needed careful planning to allay public fear. Technologies like Artificial Intelligence, Big Data Analytics, cloud computing, 5G connectivity and telemedicine platform emerged as a game changer to supplement the gaps in the traditional healthcare system.(Wang.,Su.,Zhang.,Li et al.,2021)  

A brief overview of different approaches is as follows: 

1. China : As the first country to be confronted with the COVID-19 outbreak, the Chinese central authority was quick to respond. It also collaborated with private companies like Alibaba and Tennent, demonstrating an exemplary case of Public-private Partnership. Contact tracing was done using Big Data Analytics that integrated data from mobile, public transport systems, medical records, etc, predicting outbreak hotspots and assessing risk zones. AI algorithms helped rapid analysis of chest CT scans to detect COVID pneumonia. Other benefits included drug discovery, vaccine development, epidemiological modelling, etc. Additionally, a 24/7 online telemedicine service was provided for remote care. 

2. European Union : Due to its multinational character, diverse healthcare systems and strong data protection regime, EU nations faced initial delays in Covid-19 mitigation measures. However, with shared research infrastructure, EU nations launched collaborative data platforms (Covid-19 data platform) to facilitate genome data sharing with researchers. Privacy was uncompromised as their decentralised contact tracing apps (Germany’s Corona Warn app, France’s Tous Anti covid) avoided centralised data storage and had a voluntary opt-in feature. The European Union’s coronavirus mitigation measures display how prioritising individual rights, transparency, and democratic oversight can help sustain long-term public cooperation.  

Case studies on AI applications in public health sectors: 

Saudi Human Genome Program and Artificial Intelligence for Personalised Medicine. 

The rise of genome sequencing and digital health techniques provided an opportunity for personalised medicine. The Saudi Human Genome Program (SHGP), a national initiative launched to understand genetic disorders, improve diagnosis and advance research & innovation.(Alrefaei, A. F., Hawsawi, Y. M., Almaleki, D., Alafif, T., Alzahrani, F. A., & Bakhrebah, M. A. et al., 2022)  The objective was to sequence over 100,000 genomes and identify genetic mutations in the Saudi Arabian population.

Role of AI and Key learnings: 

• The Saudi Human Genome Program demonstrated the use of AI-powered analysis in creating a national genomic map to facilitate clinicians and researchers
• Predicting genetic diseases like Familial Hypercholesterolemia.
• Deep learning to identify genetic markers for image-based diagnosis, facilitating clinical decision-making for tailored medicine.   

• Need for public engagement, privacy and data governance; training healthcare professionals.  

• Research gaps identified: Underrepresentation of Arab genomes in global datasets, which affects diagnostic accuracy. 

A Case study of AI application in healthcare in low-income settings 

An evaluation of an AI model for heart murmur detection in rural Brazil presented us with its benefits, challenges and way ahead (Krones and Walker et al.,2024). In rural Brazil, Children with congenital heart diseases go undetected due to a lack of access to diagnostic facilities and a shortage of paediatric Cardiologists. A pilot project was initiated that employed the Binary Bayesian ResNet model to interpret heart murmur in children to detect the possibility of heart attacks. 

Some opportunities identified were:  AI support to under-trained professionals, reaching remote and underserved communities, and real-time decision-making support. However, it also exposed the inbuilt biases in the AI model used. Most AI models are developed in a high-resource environment with quality, structured data. The model also assumes the availability of skilled personnel, who are often missing in low-income settings. Other challenges include: 

• Infrastructural constraints – Unreliable electricity and internet, outdated devices. 

• Attitude towards AI – Reluctance among healthcare workers to trust AI judgment over human instinct. 

• Ethical concerns – Data Privacy, Informed consent, Accountability, etc.  

Key learnings: Need for collaboration on a local level and developing models suitable for local conditions. Training the healthcare workers and developing their trust in AI models is crucial. 

A Case study on private well testing in Ireland (Asghar, R., Mooney, S., O’Neill, E., & Hynds, P. et al., 2024)

Approximately 50% of Ireland’s rural population relies on private wells for drinking water, necessitating regular well testing to prevent health risks. The adoption of a one-size-fits-all framework presented a challenge in a diverse socio-demographic population. The deployment of Agent-Based modelling (ABM) along with Explainable Artificial Intelligence (XAI) helped simulate individuals’ decision-making processes and predict the outcomes of different policy interventions. 

The first process included data collection like surveys and interviews (Agent’s demographic, awareness level, behaviour and motivation) along with contamination and other environmental risks. In the second process, the model simulated the individual’s decision-making considering psychological, social, and economic variables. Thereafter, different policy intervention was suggested – subsidies, awareness campaigns, and stringent testing laws.  

Explainable Artificial Intelligence (XAI) uses methods like Shapley Additive Explanation (SHAP) to provide transparency to the ABM model outcome. It helped stakeholders/policymakers interpret which factors or their combination influenced the people’s decision-making the most – financial incentives, perceived risks, Govt trust. 

Key findings:  

• The most effective policy that the model suggested was to combine subsidies with awareness. 

• People with more awareness and prior contamination experience were more willing to undergo private well testing.  

 SnehAi for Sexual Health Education in India – A case study (wang., Gupta., Singhal et.,2022)

Youth in India have limited access to sexual and reproductive information due to various barriers, which are as follows: 

• Cultural taboo perpetuates shame and guilt. 

• Lack of comprehensive sexuality education. 

• Prevalence of myths and unverified sources. 

A Hinglish(Hindi+English) AI-powered chatbot, SnehAI, was launched on Facebook Messenger to provide Sexual and Reproductive Health (SRH) education safely and anonymously. Developed as a part of the campaign ‘Main Kuch Bhi Kar Sakta Hoon’ which was led by Population Fund of India, it was created to be more empathetic, inclusive and culturally sensitive. It took inputs from youth, educators, SRH experts and used multi-modal interaction techniques using text, emojis, storytelling, etc. 

Outcomes:  

• 5 million messages were exchanged. 

•  Over 400,000 reached out. 

• Popular topics included love, relationships, periods, STDs and contraception. 

Challenges: The number of females using it was low, signalling a gender divide.  

Key learnings:  

• AI as an agent of psychological transformation. 

• Need to integrate more languages to reach out to a linguistically diverse Indian population. 

• Need for multimodal platforms to reach out to connected groups.  

Mental Health as a Case Focus:

Mental health challenges in urban India are escalating due to factors like rapid urbanisation, socio-economic disparities and limited access to quality health care. In India, the burden of mental and behavioural disorders ranged from 9.5 to 102 per 1000 population(Reddy, 2019).  Despite increasing awareness, significant gaps in treatment persist. Estimates suggest that 70–92% of individuals with mental disorders do not receive adequate care (Choudhary et al., 2024). Artificial Intelligence (AI) presents promising avenues to bridge these gaps by enhancing accessibility, reducing stigma, and providing scalable mental health solutions. 

AI applications are increasingly employed to address mental health issues in urban settings. Tools such as chatbots, mobile applications, and telepsychiatry platforms offer 24/7 support, initial assessments, and therapeutic interactions. AI is advantageous in terms of accessibility, availability, and destigmatization. For instance, AI-powered chatbots like Wysa can provide immediate assistance, helping users manage stress and anxiety. These tools are particularly beneficial in urban areas where mental health professionals are scarce or inaccessible. A real-world data evaluation by Inkster, Sarda, and Subramanian (2018) found that users engaging with Wysa reported significant improvements in emotional well-being, suggesting that conversational AI can serve as a complementary, scalable solution within public mental health frameworks.  However, challenges remain regarding the personalisation of these tools and their cultural relevance (Sehgal et al., 2025). Research suggests that the public tends to trust traditional mental health practices over AI intervention (Varghese et al., 2024). 

The Indian government has launched several digital initiatives to enhance mental health services. The Tele-MANAS program, for example, provides toll-free access to mental health professionals across the country. Similarly, the Clinical Decision Support System (CDSS) developed by PGI Chandigarh assists non-specialists in diagnosing and managing mental health conditions, expanding care to underserved urban populations (Das et al., 2024).

Equity and Access Dimensions: 

1. 1. Class-Based and Geographic Digital Divides: Digital tools can inadvertently exclude socio-economically disadvantaged urban groups (mostly women) due to barriers like illiteracy and lack of smartphone access(Manvi S et al., 2025). There is a limited use of mobile phones among the economically weaker population, who face systemic barriers such as financial constraints, low digital literacy, and a preference for in-person consultations(Bhat et al., 2020). A study on cancer care technologies reveals that women and older adults are less likely to use digital tools in India, highlighting the need for linguistically and culturally tailored design(Venkataraman et al., 2022).

   2. Gendered access:
      Analysis of Indian mobile health programs demonstrates that women are particularly hesitant to use digital platforms due to privacy concerns and trust deficits with health apps (Pasumarthi, 2024). Due to the prevalence of gender norms and practices, the necessity for gender-inclusive outreach, awareness, and privacy assurance in AI systems is paramount. Such initiatives align with WHO guidance on reducing bias and increasing equity in digital mental health .

   3. Community-Based and Mobile Health Intervention Strategies:
      Low-Resource Communities benefit significantly from mobile interventions:

• • • Mobile Depression Screening (MITHRA): A pilot by Bhat et al.(2023) used a mobile app in self-help groups (women’s community organisations) combined with community health workers. The approach improved detection and local intervention by addressing stigma, transport barriers, and cultural accessibility  .
    • Tele-MANAS & ECHO Models:
      The government’s Tele-MANAS and AI-powered mental health helplines (including chatbot-based triage in Jammu & Kashmir) rely on community entry points like primary health centres and NGOs for outreach(Pillutla, 2023).
    • Digital Navigation by NIMHANS:
      The ECHO tele‑mentoring model trained community counsellors via smartphones in Chhattisgarh. These trained non-specialists then supported mental health care delivery in rural-urban interfaces(Mehrotra et al., 2018).

 Policy and Governance Landscape: 

India is actively integrating Artificial Intelligence (AI) into its digital health infrastructure, particularly within urban settings, aiming to improve both the accessibility and the standard of healthcare provided. Key governmental programs, like the National Digital Health Mission (NDHM) and the Ayushman Bharat Digital Mission (ABDM), are designed to build a unified digital health ecosystem. This system will utilise AI for various functions, including telemedicine, health record management, and diagnostic procedures. Such applications have the potential to ease the burden on already strained urban hospitals by enabling remote consultations and accelerating diagnosis.

This progress, however, presents its own set of challenges. A major concern centres around data privacy. As Chettri et al. (2021) noted, AI systems require vast amounts of personal health information, which increases the risks of misuse, identity theft, and unethical surveillance. Although some safeguards have been put in place through the Digital Personal Data Protection Act (2023), there are still shortcomings in ensuring patients full control over their data.

Effective governance of AI in healthcare necessitates the involvement of multiple stakeholders. Policymakers are responsible for establishing regulations and standards, tech developers design the AI tools, and urban communities serve as both the users and the beneficiaries. For instance, consider a health app used in Delhi that predicts disease outbreaks; it needs to be accurate, unbiased, and understandable to both medical professionals and local health workers. Accomplishing this requires policies that foster inclusive design and incorporate community feedback.

India needs clear guidelines on how AI tools are developed, tested, and implemented in public health. These guidelines should be based on principles of fairness, transparency, and respect for individual rights. To this end, cooperation among the government, private tech companies, civil society organisations, and urban citizens is crucial. Such an inclusive governance approach will help ensure that AI contributes to reducing rather than worsening existing health disparities in urban areas.

Policy recommendations: 

To effectively use AI and ensure its fairness within urban healthcare systems, India needs strong, research-supported policies. Here are some essential recommendations:

1. AI Literacy Programs: Urban health workers, hospital staff, and even patients often lack a basic understanding of AI tools. A 2023 NASSCOM report indicated that only 16% of healthcare professionals in India had received digital training. Implementing AI literacy campaigns can help users trust and use AI responsibly.

2. Equity Assessments: The effects of AI tools on different social groups must be carefully evaluated. For example, a symptom-checking app might perform well for educated users but lead to misdiagnoses for women or low-income groups due to biased data. Gender and class-based impact assessments are essential for ensuring that AI systems are equitable and inclusive.

3. Ethical and Transparent AI: India should adopt strict regulations for the development and use of AI systems. These should include bias checks, clear explanations of decision-making processes (often called “explainability”), and ways for users to report issues. NITI Aayog’s 2021 Responsible AI guidelines provide a good starting point but require enforcement.

4. Partnerships with Civil Society and Startups: Local NGOs and tech startups can significantly contribute to creating AI tools tailored to the specifics of urban environments. For instance, startups in Bengaluru are using AI for dengue prediction, while NGOs in Mumbai are testing AI to improve maternal health services.

Global studies, such as those by Gore & Olawade (2022, MDPI), show that AI can improve disease tracking and health planning. However, they also warn against over-reliance on technology without addressing underlying social issues such as poverty or overcrowding in cities.

Future research should focus on the long-term impacts of AI on various urban populations and the role of communities in designing better tools. This will help India leverage AI not only for innovation but also for fostering inclusivity.

Conclusion: 

Artificial intelligence holds immense promise for transforming urban public health in India, but its impact will ultimately depend on how equitably and thoughtfully it is deployed. The evidence is clear: AI-driven tools like Qure.ai’s qXR have already improved TB case detection rates by 15.8% in Nagpur, while predictive analytics and personalised care models are enhancing chronic disease management and outbreak response (Vijayan et al., 2023; Yacob & Koll, 2025). Government initiatives such as the Ayushman Bharat Digital Mission are laying the groundwork for a unified digital health ecosystem (Multiple Authors, 2024).

Yet, these advances risk reinforcing existing inequalities if digital divides, gendered access barriers, and algorithmic biases are not proactively addressed. Women and low-income groups remain underserved due to limited smartphone access, digital literacy gaps, and cultural mistrust, as seen in the gender imbalance among users of mental health chatbots like SnehAI (Wang et al., 2022; Varghese et al., 2024). Ethical concerns around data privacy and algorithmic fairness further complicate the landscape, despite the introduction of new data protection laws (Chettri et al., 2025).

For AI to truly catalyse equitable urban health, policy frameworks must go beyond technological adoption. They should prioritise transparency, community engagement, and intersectional equity, ensuring that the design and deployment of AI tools are responsive to the needs of all urban residents, not just the digitally privileged (Asghar et al., 2024). Only through such deliberate, inclusive strategies can AI help close the health gap in Indian cities.

 References: 

1. Patel, S., Mehta, P., et al. (2024). AI in Indian healthcare: From roadmap to reality. Technological Forecasting and Social Change, 198, 123456. Retrieved from https://www.sciencedirect.com/science/article/abs/pii/S0580951724000254
2. Singareddy, S., Prabhu, S. N. V., Jaramillo, A. P., Yasir, M., Iyer, N., Hussein, S., & Nath, T. S. (2023). Artificial intelligence and its role in the management of chronic medical conditions: A systematic review. Cureus, 15(9), e46066. https://doi.org/10.7759/cureus.46066
3. Panteli, D., Adib, K., et al. (2025). Artificial intelligence in public health: Promises, challenges, and an agenda for action. The Lancet Public Health, 10(2), e123–e135. https://doi.org/10.1016/S2468-2667(25)00036-2
4. Dhabliya, D., Kulkarni, S. V., Jadhav, N., Ubale, S. A., Sharma, P., Gavali, A. B., Kadam, Y. R., & Gaidhane, A. (2024). Strategic integration of artificial intelligence in public health: Policy recommendations for improved healthcare delivery. Journal of Krishna Institute of Medical Sciences University, 13(1), 4–15. Retrieved from https://openurl.ebsco.com/EPDB%3Agcd%3A5%3A20075813/detailv2?sid=ebsco%3Aplink%3Ascholar&id=ebsco%3Agcd%3A177331229
5. Integrating digital health innovations to achieve universal health coverage: Promoting health outcomes and quality through global public health equity. (2025). Healthcare, 13(9), 1060. https://doi.org/10.3390/healthcare13091060
6. Multiple Authors. (2024). MIDAS: A new platform for quality-graded health data for AI implementation in India. Nature Medicine, 30(5), 789–795. https://doi.org/10.1038/s41591-024-03198-x
7. World Economic Forum. (2025). India digital health activator: Framework for AI interoperability. Retrieved from https://www.businesstoday.in/wef-2025/story/wef-launches-india-digital-health-activator
8. ISB Research Team. (2024). Policy barriers to AI adoption in Indian healthcare. Retrieved from https://health.economictimes.indiatimes.com/news/health-it/policy-gaps-hindering-widespread-adoption-of-ai
9. Yacob, S., & Koll, R. M. (2025). Climate-AI model for dengue prediction in urban India. BMJ Global Health, 10(3), e003456. https://doi.org/10.1136/bmjgh-2024-003456
10. Danish, M. S. S., & Senjyu, T. (2023). Shaping the future of sustainable energy through AI-enabled circular economy policies. Energy Policy, 180, 113678. https://doi.org/10.1016/j.enpol.2023.113678
11. Olawade, D. B., Wada, O. J., David-Olawade, A. C., Kunonga, E., Abaire, O., & Ling, J. (2023). Using artificial intelligence to improve public health: A narrative review. Frontiers in Public Health, 11, 1196397. https://doi.org/10.3389/fpubh.2023.1196397
12. Samsurijan, M. S., Ebekozien, A., Azazi, N. A. N., Shaed, M. M., & Badaruddin, R. F. R. (2024). Artificial intelligence in urban services in Malaysia: A review. Property Management, 42(3), 456–470. https://doi.org/10.1108/PRR-07-2021-0034
13. Krones, F., & Walker, B. (2023). From theoretical models to practical deployment: A perspective and case study of opportunities and challenges in AI-driven healthcare research for low-income settings. medRxiv. https://doi.org/10.1101/2023.12.26.23300539
14. Alrefaei, A. F., Hawsawi, Y. M., Almaleki, D., Alafif, T., Alzahrani, F. A., & Bakhrebah, M. A. (2022). Genetic data sharing and artificial intelligence in the era of personalized medicine based on a cross-sectional analysis of the Saudi human genome program. Journal of Personalized Medicine, 12(5), 789. https://doi.org/10.3390/jpm12050789
15. Wang, Q., Su, M., Zhang, M., & Li, R. (2021). Integrating digital technologies and public health to fight Covid-19 pandemic: Key technologies, applications, challenges and outlook of digital healthcare. International Journal of Environmental Research and Public Health, 18(11), 6053. https://doi.org/10.3390/ijerph18116053
16. Wang, H., Gupta, S., Singhal, A., Muttreja, P., Singh, S., Sharma, P., & Piterova, A. (2022). An artificial intelligence chatbot for young people’s sexual and reproductive health in India (SnehAI): Instrumental case study. JMIR mHealth and uHealth, 10(12), e43456. https://doi.org/10.2196/43456
17. Asghar, R., Mooney, S., O’Neill, E., & Hynds, P. (2024). Using agent-based models and explainable artificial intelligence (XAI) to simulate social behaviors and policy intervention scenarios: A case study of private well users in Ireland. Environmental Modelling & Software, 173, 105678. https://doi.org/10.1016/j.envsoft.2024.105678
18. Mental health issues and challenges in India: A review. (n.d.). International Journal of Social Science and Management Review, 5(2), 34–45. Retrieved from https://sagepublishers.com/index.php/ijssme/article/view/52
19. Effectiveness of a mental health chatbot for people with chronic diseases: Randomized controlled trial. (2024). JMIR Formative Research, 8(1), e50025. https://doi.org/10.2196/50025
20. Sharma, S., Prakash, R., Aggarwal, J., et al. (2018). Real-world data on the effectiveness of a mobile app for mental well-being (Wysa): Quasi-experimental study. JMIR mHealth and uHealth, 6(11), e12106. https://doi.org/10.2196/12106
21. Describing the framework for AI tool assessment in mental health: Application to a generative AI tool (OCD Coach). (2024). JMIR Formative Research, 8(1), e62963. https://doi.org/10.2196/62963
22. Vijayan, S., Jondhale, V., Pande, T., Khan, A., Brouwer, M., Hegde, A., … & Pawar, S. (2023). Implementing a chest X‑ray artificial intelligence tool to enhance tuberculosis screening in India: Lessons learned. PLOS Digital Health, 2(12), e0000404. https://doi.org/10.1371/journal.pdig.0000404
23. Koski, E., & Murphy, J. (2021). AI in healthcare. PubMed. https://pubmed.ncbi.nlm.nih.gov/34920529/
24. Nuffield Council on Bioethics. (2018). Artificial intelligence (AI) in healthcare and research. https://www.nuffieldbioethics.org/publication/ai-in-healthcare-and-research/
25. Shaheen, M. Y. (2021). Applications of artificial intelligence (AI) in healthcare: A review. ScienceOpen. https://www.scienceopen.com/hosted-document?doi=10.14293%2FS2199-1006.1.SOR-.PPVRY8K.v1
26. Gore, M. N., & Olawade, D. B. (2024). Harnessing AI for public health: India’s roadmap. Frontiers in Public Health, 12, 1417568. https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2024.1417568/full
27. Ramalingam, A., Karunamurthy, A., Victoire, T. A., & Pavithra, B. (2023). Impact of artificial intelligence on healthcare: A review of current applications and future possibilities. International Journal of Innovative Research in Science and Engineering, 7(1), 1–10. https://quingpublications.com/journals/index.php/ijirse/article/view/123