Beyond forecasting: Can AI help Indian cities fight heatwave better?

AI is moving beyond heatwave forecasts into urban planning, healthcare and power management, but experts warn gaps in data, governance and access still remain

Indian weather agencies are increasingly turning to artificial intelligence (AI) to sharpen heatwave forecasting and strengthen early-warning systems. So far, much of the effort has focused on improving weather models that can predict when temperatures are likely to cross dangerous thresholds and help authorities prepare in advance.

 

But as heatwaves grow more frequent, prolonged, and deadly, experts say the conversation is beginning to shift beyond forecasting alone and towards whether AI can also help cities manage extreme heat more effectively on the ground.

 

The question has become increasingly urgent as India faces recurring heat stress, especially in urban centres. The India Meteorological Department (IMD) has issued a red and severe heatwave alert across large parts of northwest, central, and eastern India, warning that extreme temperatures are likely to continue until May 27-28. The ongoing pre-monsoon heat spell has already pushed temperatures beyond 45 degrees Celsius in several regions this month.

   

The health toll is also becoming visible. Earlier this month, Telangana reported at least 16 heatstroke-related deaths, while Andhra Pradesh recorded a rise in heat-linked illnesses as prolonged exposure to extreme temperatures continued to affect vulnerable populations.

 

This is where AI’s next role is being tested: not just predicting heat, but helping cities decide what to do before it turns deadly.

From predicting heat to managing it

“Forecasting is only the first layer of heat resilience,” said Vijay Sampathkumar, chief business officer at Refroid Technologies, which designs cooling solutions for AI, HPC, and data centers. “The larger opportunity lies in how cities operationalise those insights across infrastructure, healthcare, and public services.”

 

Experts say AI systems are increasingly being used to identify urban heat islands, improve cooling infrastructure planning, and strengthen healthcare preparedness. One of the biggest developments has been the rise of hyperlocal heat mapping, where AI analyses satellite imagery, land-use patterns, vegetation cover, and building density to identify vulnerable neighbourhoods at street or building level.

 

“Beyond forecasting, AI’s most important contribution is converting hazard information into precise local actions that reduce harm,” said Professor Anjal Prakash, faculty of public policy at FLAME University.

 

He said high-resolution heat and vulnerability maps can help authorities prioritise interventions such as tree planting, cool roofs, reflective pavements, and shaded public spaces as Indian cities grapple with worsening urban heat island effects.

 

Agendra Kumar, managing director of Esri India, said GIS and GeoAI systems were increasingly helping authorities identify where such interventions could create the highest cooling impact.

 

“GIS models can overlay land surface temperature data with green cover and building density to pinpoint exactly where a cool roof, tree corridor, or ventilation pathway can deliver the highest cooling impact,” Kumar said.

 

The role of AI is also expanding into healthcare preparedness and utility management.

 

“In the health sector, AI can fuse climate data with demographic and health facility information to flag wards at the highest risk of heat-related illness,” Prakash said, adding that such systems could help authorities pre-position ambulances, strengthen hospital preparedness, and deliver targeted public-health messaging before temperatures peak.

 

Electricity management is emerging as another critical area. India’s peak power demand touched a record 270.82 gigawatts (GW) during daytime solar hours on May 21, as soaring temperatures drove up cooling requirements across several regions.

 

According to Vasudha Madhavan, founder and chief executive officer of climate-tech investment banking firm Ostara Advisors, AI systems are increasingly helping cities forecast cooling demand and anticipate stress on electricity grids before blackouts occur.

 

“The larger shift is from reactive response to proactive, data-driven heat resilience planning,” she said.

 

Experts also pointed to emerging technologies such as digital twins — virtual replicas of urban systems that allow authorities to simulate heat stress scenarios before implementing interventions. These systems can help cities test the impact of urban greening, reflective infrastructure, and shaded corridors before investing public resources.

The challenge beyond metros

While most AI-led climate systems are currently concentrated in larger cities, experts say the next challenge lies in scaling these technologies beyond metropolitan India.

 

“Scaling climate technologies to tier-II, tier-III, and rural regions is both a technological and governance challenge,” Sampathkumar said.

 

Weak internet connectivity, inconsistent climate datasets, sparse sensor networks, and limited institutional capacity continue to remain major barriers in smaller towns and rural areas. This has pushed firms towards low-bandwidth systems, mobile-first platforms, and local-language alerts.

 

Prakash said several organisations were also exploring multilingual SMS alerts, offline-capable applications, and community-based dissemination systems involving ASHA workers, panchayats, and local radio networks.

 

The challenge becomes even more complex when viewed through the lens of digital inequality.

 

Janhavi Bhujabal, consultant at the Climate and Sustainability Initiative, warned that AI-led climate systems risked deepening existing inequalities if they fail to account for vulnerable populations and weak last-mile connectivity.

 

“Information disseminated to a migrant labourer from Tamil Nadu or West Bengal in the Hindi heartland, in his or her non-native language, risks both ex-ante and ex-post delays in response to early warning alerts and institutional support,” she said.

Technology alone cannot solve the problem

Despite rapid advances in AI systems, experts cautioned that technology by itself cannot resolve India’s heat vulnerability.

 

Anupam Shrey, founder of AI climate risk assessment firm Plutas.ai, said the biggest barriers were not purely technical.

 

“First, governance accountability. AI cannot operate in a policy vacuum,” Shrey said, pointing out that many heat action plans still lacked detailed ward-level risk mapping.

 

He also highlighted what he called the “indoor heat blind spot”, particularly in informal settlements where indoor temperatures can remain dangerously high even after sunset.

 

“Healthcare expenses surge, productivity collapses, and well-being deteriorates, yet indoor heat in informal housing remains systematically undercounted in official risk assessments,” he said.

 

Experts also warned against over-reliance on AI systems in regions where data quality and infrastructure remain weak.

 

“The specific danger in heat resilience contexts is false precision,” Shrey said. “A model that outputs a neighbourhood-level heat risk score with high apparent confidence, but is trained on sparse or biased data, can actively misdirect resource allocation.”

 

Prof Prakash added that dependence on cloud connectivity and digital infrastructure could itself become a vulnerability during crises.

 

“Power or network outages often coincide with crises, and an AI-centric system can fail precisely when it is most needed,” he said.

 

Bhujabal argued that policymakers should treat AI-led climate resilience systems as a public good rather than purely technology projects.

 

“Policymakers should instead treat them as a public good that combines climate science, governance, local language delivery, internet access, and financing challenges,” she said.

 

Globally, several cities are already experimenting with AI-assisted heat adaptation - from AI-powered tree canopy planning and cool-roof deployment in US cities to district-level heat mapping systems in India. Still, experts say the larger challenge now is not whether AI can generate climate intelligence. It increasingly can. The harder question is whether Indian cities can build the governance systems, infrastructure, and institutional coordination needed to turn that intelligence into meaningful protection on the ground.