The present invention provides a method of providing security for a machine learning, ML, classifier against adaptive gray-box adversarial samples. According to embodiments, the method comprises inspecting, by a detector component, input samples submitted to the ML classifier to identify input samples that appear as noisy versions of previously submitted queries and/or as highly distorted input samples; and forwarding, by the detector component, only those input samples to the ML classifier that were not identified as noisy versions of previously submitted queries or as highly distorted input samples.