AI techniques, from machine learning (ML) to the most recent developments in deep learning (DL), are widely utilized in optical communication and networks. This chapter explores advanced ML and DL methods and their applications in processing fiber optic sensors. It emphasizes the potential of these algorithms for the design and development of intelligent sensors. The chapter addresses the challenges and limitations of fiber optic sensors and how AI has addressed these issues. AI has significantly enhanced signal processing in optical fiber sensors by improving data analysis, increasing accuracy, and accelerating decision-making. AI techniques facilitate the effective management and interpretation of complex sensor data, resulting in more accurate and timely insights. Recent advancements in ML-based fiber sensors are reviewed, highlighting the problems encountered with conventional methods. Machine learning (ML) techniques have been applied to optimize various aspects of fiber optic sensors. In the past 20 years, the integration of machine learning (ML) and artificial intelligence (AI) has transformed various scientific fields, facilitated by tools like MATLAB, Scikit-Learn, and Tensor Flow. These advancements include Scientific Machine Learning (SciML) and physics-informed ML algorithms, which combine domain knowledge with data-driven models. Deep learning has notably improved pattern recognition, often matching human-level performance. This chapter highlights recent developments in ML and AI that enhance fiber optic sensing (FOS) applications and provide a thorough overview of ML techniques applied to optical fiber sensors focusing on various AI methods that enhance sensor performance. Additionally, it discusses future developments in this evolving research field.