This paper discusses the use of intelligent system approaches, a combination of fuzzy logic and artificial neural network (ANN) techniques, for the automated inspection of commercial carbon/carbon (C/C) composite aircraft brake disks. The hybrid fuzzy neural approach was applied to the infrared thermography and air-coupled ultrasonic testing (UT) results obtained for C/C composite samples. This approach in detecting and classifying discontinuities follows a general pattern recognition problem that is based on three steps: segmentation, feature extraction and classification. To achieve this, discontinuous regions were first segmented from the non-discontinuous regions. After that, a 3 × 3 processing window was considered within the infrared thermography and air-coupled UT results. Then, five linguistic labels for the center value and mean value of the 3 × 3 processing window were generated as: very low, low, medium, high and very high. Finally, a Mamdani inference was applied for the fuzzification, rule evaluation and aggregation, and defuzzification processes. It was demonstrated that the 25 fuzzy rules developed were able to identify and classify the discontinuous areas in the C/C samples. It was also shown that this approach greatly reduces the noise, leading to better discontinuity recognition. After the successful implementation of fuzzy logic, an ANN was used for automatic discontinuity detection and recognition in C/C composites. A multi-layer perceptron neural network (MLP) with a scaled conjugate back-propagation learning algorithm was used to train the network in an offline mode. Numerous MLPs were repeatedly trained by changing the network size, structure and number of hidden neurons, and by varying different learning, training and activation functions. The network training was conducted until a steady state was reached, where there was no further change in the synaptic weights.