AI Accountability Project Launched as Delhi-NCR Confirms Winter Smog Emergency

AI Accountability Project Launched as Delhi-NCR Confirms Winter Smog Emergency

Aether 360 aims to bridge the gap between environmental data and direct health consequences by developing the world’s first AI model capable of calculating the “Attribution Rate” (A-Rate).

RMN News Environment Desk
December 7, 2025

The Delhi-NCR region is grappling with a severe public health emergency, having experienced one of the most polluted months in recent years during November 2025. Data indicates that all 10 of India’s most polluted cities were located in the Delhi-NCR region during that month.

Amid this crisis, journalist and activist Rakesh Raman launched the self-funded Aether 360 project, a Proof-of-Concept (PoC) designed to create accountability where traditional policy has failed and to counter official denial regarding pollution-related health crises.

The Scale of the Crisis

Ghaziabad topped the national pollution charts in November 2025, recording a monthly PM2.5 average of 224 micrograms per cubic metre (µg/m³), which is more than three times the national safe limit. The city experienced 19 ‘very poor’ days and 10 ‘severe’ days, with every day exceeding the National Ambient Air Quality Standards (NAAQS). The national capital ranked fourth, recording a monthly average PM2.5 level of 215 µg/m³, double its October concentration.

Experts from the Centre for Research on Energy and Clean Air (CREA) noted that despite a significant reduction in stubble-burning influence (which averaged only 7% compared to 20% the previous year), pollution levels were higher in 20 out of 29 NCR cities. This clearly suggests that dominant drivers are year-round sources such as transport, industry, power plants, and other combustion sources. Without sector-specific emission cuts addressing these drivers, cities will continue to breach standards.

A New Metric for Health Attribution

Aether 360 aims to bridge the gap between environmental data and direct health consequences by developing the world’s first AI model capable of calculating the “Attribution Rate” (A-Rate).

The A-Rate is a metric specifically designed to quantify the exact probability that a specific patient’s acute respiratory or cardiac hospital admission was directly caused by a recent spike in air pollution. To establish this causal connection, the tool employs Explainable AI (XAI) to generate a “Pollution Probability Link” (PPL) between granular air quality readings and de-identified patient data.

The goal of providing this quantifiable, localized, causal evidence is to break the policy stalemate and justify robust public health interventions. Aether 360’s developers anticipate that precisely counting every hospital admission caused by toxic air would force a profound change in policy. Critical to the project’s success, however, is securing institutional partnerships with organizations such as the WHO, UNEP, or AIIMS to validate the model using real-world, anonymized patient data.

If the high pollution levels are like a silent, invisible assailant repeatedly attacking a community, the Aether 360 project acts like a high-tech forensic detective, using AI to finally identify and precisely measure the link between the assailant (pollution spikes) and specific crime victims (hospital admissions), providing the clear evidence needed to demand institutional intervention.