We investigate experimentally a novel model-free controlstrategy, called Machine Learning Control (MLC), foraerodynamic drag reduction of a 3D bluff body. Fluidicactuation is applied at the blunt trailing edge of the bodycombined with a curved deflection surface. The impact ofactuation on the flow is monitored with base pressure sensors. The applied model-free control strategy detects andexploits nonlinear actuation mechanisms in an unsupervised manner with the aim of minimizing the drag. Keyenabler is linear genetic programming as simple and efficient framework for systems with multiple inputs (actuators) and multiple outputs (sensors). The ansatz of control laws include periodic forcing, multi-frequency forcing and sensor-based feedback control. Approximately33% base pressure recovery is achieved by the optimalcontrol law for a turbulent flow at Re_H ≈ 3 Ã 10 5 basedon body height.