The advent of artificial intelligence (AI) adoption has stirred new application in financial markets based on the use of AI. One application is exploration of AI techniques to predict risk asset prices [1]. Traditional theories in finance such as the Efficient Market Hypothesis (EMH) [2] and the random walk theory suggest that stock prices are inherently unpredictable due to their dependence on new information [3]. But, emerging research in behavioral finance theories such as the Noise Trader Approach (NTA) and the fractal market hypothesis challenges these views. Behavioral finance theories highlight the limitations of classical finance principles especially by emphasizing the role that irrational investor behavior plays in the market dynamics [4]. Despite the growing traction of the behavioral finance based theories, there is a lack of representation of these approaches in AI-based financial models. This research study, by integrating social sentiment analysis-based behavioral finance theory into AI algorithms for asset price prediction, attempts to bridge this gap [4]. Specifically, the research focuses on the application of the social sentiment based AI algorithms with cryptocurrency based illustrations to investigate applicability of behavioral finance concepts. The aim is to explore if the integration of behavioral finance based theories and classical finance theories can improve the accuracy of AI-based forecasting. This study provides a novel perspective on the market predictability and challenges the long-held assumptions of classical finance. This study contributes to the scholarly discourse by augmenting behavioral finance theories such as social-sentiment analysis with traditional financial forecasting techniques thus enhancing the applicability of AI. The findings, by developing more advanced methods for forecasting risk asset prices, aim to benefit academic researchers and financial practitioners.