Impact of artificial intelligence and energy-technology R&D on energy transition in the United States

As artificial intelligence reshapes the global economy, a critical question emerges: can digital transformation accelerate the renewable energy transition, or will it intensify pressure on clean-energy systems? To this end, this study explores the drivers of renewable energy consumption using a wavelet quantile-based approach and quarterly data from 1980Q1 to 2021Q4. The results show that artificial intelligence mostly exerts a negative effect on renewable energy consumption, especially in the short and long run. Energy-technology R&D improves renewable energy consumption mainly at higher quantiles in the short run, while its medium- and long-term effects are mixed. Financial development provides the strongest positive support, particularly at lower and higher quantiles. Trade globalization has mixed short-run effects but becomes more supportive in the medium term through technology transfer, renewable-energy imports, and clean supply chains. In contrast, economic growth mostly reduces renewable energy consumption in the medium and long run when tied to conventional energy structures. Robustness checks confirm that WQQR captures richer distributional patterns than WQR, while wavelet Granger causality reveals strong feedback linkages among the variables. Building on the empirical evidence, this study offers targeted and actionable policy recommendations.