The improvement of remote sensing systems, together with the emergence of new model-fitting algorithms based on machine-learning techniques, has allowed the determination of the partial pressure of carbon dioxide (<em>p</em>CO<sub>2,sw</sub>) and pH (pH<sub>T,sw</sub>) in the waters of the Canary Islands. Among all the fitted models, the most powerful one was the bootstrap aggregation (<em>bagging</em>), giving a RMSE of 2.0 µatm (R<sup>2</sup> > 0.99) for <em>p</em>CO<sub>2,sw</sub> and RMSE of 0.002 for pH<sub>T,is</sub>, although the multilinear regression (<em>MLR</em>), neural network (<em>NN</em>) and categorical boosting (<em>catBoost</em>) also have a good predictive performance, with RMSE ranging from 5.4 to 10 µatm for 360 < <em>p</em>CO<sub>2,sw</sub> < 481 µatm and from 0.004 and 0.008 for 7.97 < pH<sub>T,is</sub><sup> </sup>< 8.07. Using the most reliable model, it was determined that there is an interannual trend of 3.51 ± 0.31 µatm yr<sup>-1</sup> for <em>p</em>CO<sub>2,sw</sub> (which surpasses the rate of increase for atmospheric CO<sub>2</sub> of 2.3 µatm yr<sup>-1</sup>) and an increase in acidity of -0.003 ± 0.001 pH units yr<sup>-1</sup>. The increase in both, the atmospheric CO<sub>2</sub> and the sea surface temperature of 0.2 °C yr<sup>-1 </sup>observed in the 6-year period, influenced by the unprecedented 2023 marine heat wave, contribute to this important rate. Considering the Canary Islands between 13°–19° W and 27°–30° N, the region has moved from a slight CO<sub>2</sub> source of 0.90 Tg CO<sub>2</sub> yr<sup>-1</sup> in 2019 to 4.5 Tg CO<sub>2</sub> yr<sup>-1</sup> in 2024. After 2022, eastern locations that acted as an annual sink of CO<sub>2</sub> switched to acting as a source.