<![CDATA[The rapid proliferation of cloud-based information systems has introduced unprecedented cybersecurity challenges, necessitating robust and adaptive intrusion detection mechanisms. This paper proposes a novel Blockchain-Enabled Artificial Intelligence Framework for Intrusion Detection (BAIFD) in cloud environments. The proposed framework integrates a federated deep learning architecture with immutable blockchain ledger technology to achieve decentralized, tamper-resistant, and highly accurate threat identification. Two formal models are presented: (i) a Federated Threat Detection Model (FTDM) that coordinates distributed AI agents across heterogeneous cloud nodes without sharing raw data, and (ii) a Blockchain Consensus Validation Model (BCVM) that ensures the integrity and provenance of threat intelligence records. Extensive experiments conducted on three benchmark datasets: NSL-KDD, CICIDS2017, and UNSW-NB15 demonstrate that BAIFD achieves a detection accuracy of 99.1%, a false-positive rate of 0.43%, and an average latency of 18.7 ms, outperforming seven state-of-the-art baselines. Six architectural and analytical figures and five comparative performance tables are provided to illustrate the framework design, model workflows, and evaluation results. The findings confirm that the convergence of blockchain and federated deep learning delivers a scalable, privacy-preserving, and computationally efficient solution for next-generation cloud intrusion detection systems.]]>
