Abstract Slickwater fluids, used to undertake fracturing in low-permeability reservoirs, may be derived from produced water consisting of a range of dissolved salts. The fluids are pumped downhole at high flowrates, and hence friction reducers, e.g., anionic polyacrylamides (APAMs) are added, which also impart viscosity to the fluid resulting in better proppant transport. The present work investigates the effect of an APAM copolymer on the viscosity of slickwater fluids; specifically, at high salinity and hardness conditions. The experimental part of this study demonstrated the impact of the parameters—APAM concentration and salt type/concentration—on slickwater fluid viscosity. In a freshwater–APAM fluid, as monovalent salt (salinity) is added incrementally, fluid viscosity decreased initially owing to the charge-shielding effect; and, then viscosity values were leveled off beyond a certain salinity level. However, a very peculiar behavior was observed for the addition of divalent salts (hardness) to fluid systems. Initially, as hardness increased up to 50k ppm (parts per million), as expected, fluid viscosity showed a significant decrease; on the contrary, as the hardness was raised beyond 50k ppm, the solution viscosity showed a distinctive increase up to 250k ppm. This phenomenon may be explained on the basis of the reverse charge-shielding effect, i.e., excessive divalent ionicity inducing repulsion between polymer charge sites. To model the above experimentally observed non-monotonous viscosity behavior, various machine learning models were employed; support vector regression (SVR)based models predicted the slickwater fluid viscosity with maximum accuracy. Sensitivity analysis was carried out to determine the prominence of the studied input parameters. The modeling work would assist in minimizing trial-and-error in designing/optimizing a slickwater fluid system.