AI for Social Good: An Investigation of the Causal Relationship Between Environmental Regulations and Their Effects on Air Pollution in London, UK
A new study published on arXiv details a Bayesian deep learning framework designed to assess the impact of environmental regulations on air pollution in London. The model, a Bayesian LSTM, integrates various data sources including PM$_{2.5}$ concentrations, meteorological data, socioeconomic indicators, and policy implementation dates. Researchers used this framework to estimate that London's regulations led to an average reduction of 1.88 $\mu$g/m$^3$ in PM$_{2.5}$ levels between 2010 and 2020, with the effects becoming more pronounced after 2013. AI
IMPACT Demonstrates how causal AI can support environmental accountability and evidence-based governance.