Taşkın, DilvinKim, EonsooMukhtarov, ShahriyarKirikkaleli, DervişKılıç Depren, SerpilPark, JinsuDepren, Serpil KilicKartal, Mustafa Tevfik2026-04-072026-04-0720250301-47971095-863010.1016/j.jenvman.2025.1279242-s2.0-105020981745https://hdl.handle.net/123456789/14141https://doi.org/10.1016/j.jenvman.2025.127924Due to the increasing negative effects on humanity, searching for potential solutions to combat environmental problems has been developing. Accordingly, the study examines the effect of a set of critical factors on environmental sustainability (ES) proxied by ecological footprint (EFP) and load capacity factor (LCF) in China. In this context, the study considers AI-related patents, energy transition, environmental policy stringency (EPS), income, and energy consumption (EC) sub-types and applies the Kernel Regularized Least Squares (KRLS) approach on data from 2000 to 2020 within the context of marginal effect analysis. The outcomes show that (i) AI-related patents and energy transition are completely ineffective to ensure ES; (ii) EPS are marginally effective only at 0.25th and 0.75th percentiles to support ES; (iii) economic growth as well as oil, gas, and coal EC are not good for ES across all percentiles; (iv) nuclear EC is only helpful at 0.25th percentiles, whereas renewable EC is completely unbeneficial; (v) KRLS approach presents successful prediction outcomes around 99.7 % (vi) some variables (i.e., nuclear and renewable EC as well as EPS); have marginal and varying effects across percentiles, whereas some others have not. Thus, the study empirically demonstrates the inefficiency of AI-related patents and energy transition on the ES, whereas EPS and nuclear EC can be helpful to develop ES in the Chinese case.eninfo:eu-repo/semantics/closedAccessIncomeAI-Related PatentsEnvironmental SustainabilityChinaEnvironmental Policy StringencyEnergy TransitionEnergy ConsumptionArticle