In the current research endeavor, the predictive model of tool wear in rotary tool micro-ultrasonic machining (RTMUSM) process using artificial intelligence has been reported for the first time. An adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) approaches are adopted for modeling of total volumetric wear (TVW) in RTMUSM process. The models were developed after conducting the exhaustive experimentation. Six RTMUSM input process parameters (i.e. abrasive size, rotation speed, depth of channel, slurry concentration, feed rate, and power rating) were selected as input data with TVW of tool as output data. The developed model has been validated through experimental results. Further, statistical analysis by calculating variance account for (VAF), coefficient of correlation (R-value), mean absolute percentage error (MAPE), and root mean square error (RMSE) was carried out to estimate the potential of the models. The results obtained by statistical analysis proclaimed that the performance of ANFIS model (R-value = 0.9966, MAPE = 1.26%, RMSE = 1.11%, and VAF = 99.94) was superior as compared to ANN model and it can be used to measure the TVW of tool in RTMUSM. Furthermore, genetic algorithm based optimization was performed and required wear compensation of tool is recommended.