The chapter aims to discuss and justify the feasibility and usefulness of integrating MEMS sensors and Artificial Intelligence (AI) to enhance the quality of sensor signals and improve, as a consequence on other performance aspects of sensing systems. The focus is on a particular AI tool, i.e. Artificial Neural Networks (ANNs), whose non-linear mapping abilities are exploited. A brief introduction to ANN techniques used generally in non-linear systems identification and control is provided with the purpose of illustrating the properties which make Neural Networks (NNs) suitable for integration with MEMS sensors, from a systemic viewpoint. Further, a capacitive accelerometer is considered as a case study and a methodology for identifying and controlling the sensor is presented. The material in this chapter forms a reference point for the review of some of the successful NN based sensor systems designs presented in Chapter 8. The chapter does not treat the implementation implications of an AI based sensor system design approach, as the view taken is that the system will contain the means to digitise the sensed signals and has the computation abilities to process them.