The number of studies on the relationship between technological innovation and CO2 emissions has gradually increased in recent years, although there is no clear agreement in the literature. Previous research has revealed both positive and negative consequences of technological innovation on the environment. Moreover, most researchers have used linear approaches to explore this connection, which can result in spurious outcomes when nonlinearities exist in the data. According to this background, this research utilizes asymmetric ARDL and spectral causality approaches to assess the asymmetric connection between technological innovation and CO2 emissions in Sweden utilizing data from 1980 to 2018. In addition, the disaggregated asymmetric effects of technological innovation (patent resident and patent nonresident) on CO2 are also captured in this study. The Nonlinear Autoregressive Distributed lag (NARDL) results showed that positive (negative) shocks in economic growth enhance environmental quality in Sweden. Furthermore, a positive (negative) shock in technological innovation causes a decrease (increase) in CO2. Similarly, a positive (negative) shock in patent nonresident and residents leads to a decrease (increase) in CO2 emissions in Sweden. The outcomes from the spectral causality revealed that in the medium and long term, aggregate and disaggregate technological innovation can predict CO2 emissions in Sweden. This study has significant policy implications for policymakers and the government in Sweden. Based on these findings, the study suggests that the government of Sweden should investment in technological innovation since it plays a vital role in curbing environmental degradation.