The Dark Energy Spectroscopic Instrument (DESI) collaboration has released measurements of baryon acoustic oscillations (BAO) from the first year of observations, which, when combined with cosmic microwave background (CMB) and type Ia supernova data, indicate a preference for time-evolving dark energy. To evaluate the robustness of this preference, the authors replace DESI BAO measurements at \(z < 0.8\) with SDSS BAO measurements in a similar redshift range. Using the \(w_0 w_a\) CDM model, they find that the evolution of dark energy parameters is consistent with the \(\Lambda\)CDM model. The analysis of \(z^2\) statistics across various BAO datasets shows that the DESI's preference for evolving dark energy is primarily driven by the two LRG samples at \(z_{\text{eff}} = 0.51\) and \(z_{\text{eff}} = 0.71\), with the latter having the most significant impact. Taking this preference seriously, the authors study a general Horndeski scalar-tensor theory, which provides a physical mechanism to cross the phantom divide (\(w = -1\)). Using the Effective Field Theory (EFT) of dark energy and the \(w_0 w_a\) CDM background cosmological model, they derive constraints on the parameters \(w_0\) and \(w_a\), showing good consistency with the standard \(w_0 w_a\) CDM model. The modified gravity model shows a preference over \(\Lambda\)CDM at the \(2.4\sigma\) level, while for \(w_0 w_a\) CDM it is at \(2.5\sigma\). They conclude that modified gravity offers a viable physical explanation for DESI's preference for evolving dark energy.The Dark Energy Spectroscopic Instrument (DESI) collaboration has released measurements of baryon acoustic oscillations (BAO) from the first year of observations, which, when combined with cosmic microwave background (CMB) and type Ia supernova data, indicate a preference for time-evolving dark energy. To evaluate the robustness of this preference, the authors replace DESI BAO measurements at \(z < 0.8\) with SDSS BAO measurements in a similar redshift range. Using the \(w_0 w_a\) CDM model, they find that the evolution of dark energy parameters is consistent with the \(\Lambda\)CDM model. The analysis of \(z^2\) statistics across various BAO datasets shows that the DESI's preference for evolving dark energy is primarily driven by the two LRG samples at \(z_{\text{eff}} = 0.51\) and \(z_{\text{eff}} = 0.71\), with the latter having the most significant impact. Taking this preference seriously, the authors study a general Horndeski scalar-tensor theory, which provides a physical mechanism to cross the phantom divide (\(w = -1\)). Using the Effective Field Theory (EFT) of dark energy and the \(w_0 w_a\) CDM background cosmological model, they derive constraints on the parameters \(w_0\) and \(w_a\), showing good consistency with the standard \(w_0 w_a\) CDM model. The modified gravity model shows a preference over \(\Lambda\)CDM at the \(2.4\sigma\) level, while for \(w_0 w_a\) CDM it is at \(2.5\sigma\). They conclude that modified gravity offers a viable physical explanation for DESI's preference for evolving dark energy.