The rapid proliferation of Internet of Things (IoT) devices has introduced significant security challenges, including vulnerability to cyberattacks, inadequate authentication mechanisms, and concerns over data integrity and privacy. This research proposes an adaptive blockchain-based security framework tailored to address these issues by utilizing the decentralized, tamper-resistant properties of blockchain technology. The framework integrates a hybrid blockchain architecture-combining permissioned and public chains-to ensure both scalability and transparency across diverse IoT environments. Lightweight SIMple Optimized Notation (SIMON) cryptographic algorithms are employed to secure device-level communications and authentication processes, offering efficiency suitable for resource-constrained IoT devices. For secure and scalable communication, the framework utilizes the Message Queuing Telemetry Transport (MQTT) protocol, enabling lightweight and reliable messaging between IoT nodes and the blockchain network. Smart contracts are deployed for automated enforcement of security policies, while a machine learning component based on Long Short-Term Memory (LSTM) networks is used for real-time anomaly detection, enabling dynamic response to evolving threats. Simulation tests in smart home and industrial IoT environments demonstrate the framework’s potential to enhance security, privacy, and resilience. The study also evaluates the trade-offs among security, latency, and energy consumption, offering practical insights into deploying blockchain-enabled and machine learning-augmented solutions within IoT ecosystems.