Cloud storage has emerged as the backbone of modern digital ecosystems, enabling seamless data access, sharing, and collaboration across individuals, enterprises, and government organizations. However, the centralized nature of conventional cloud architectures makes them vulnerable to critical security challenges such as data breaches, manipulation, unauthorized access, and single-point failures. To address these issues, this study proposes a hybrid intelligent cloud security framework that integrates Artificial Intelligence (AI) and Blockchain technologies. Blockchain ensures decentralized trust through cryptographic immutability, distributed consensus, and smart contracts that automate data access and policy enforcement without third-party intervention. Simultaneously, AI specifically Long Short-Term Memory (LSTM) networks is employed for anomaly detection, analysing user activity logs and behavioural patterns to identify irregularities or potential intrusions in real time. The system dynamically adjusts resource allocation and access privileges based on AI-driven insights, enhancing operational efficiency and security adaptability. Experimental evaluation demonstrates that the model achieves high performance in terms of accuracy, precision, recall, F1-score, latency, and throughput, validating its robustness and scalability under varying network conditions. By combining AI's predictive intelligence with blockchain's decentralized integrity, the proposed approach delivers a secure, transparent, and self-optimizing cloud storage framework suitable for data-sensitive domains such as healthcare, finance, e-governance, and smart industries.