In modern information societies, the widespread use of data sharing platforms has facilitated information exchange and collaboration among users, but the risk of sensitive information leakage has become increasingly severe. ABE (Attribute-based encryption) is employed to address data sharing challenges in multiuser and distributed systems. However, verifying the authenticity of the attribute sets provided by users remains an issue to be resolved. Blockchain, with its decentralized and tamper-proof characteristics, offers stronger privacy protection. Additionally, the dynamic changes in users’ permission attribute sets and the efficiency of retrieving and updating these attributes during the data-sharing process remain open problems. This paper proposes a blockchain-based fine-grained access control framework. To ensure the authenticity of users’ attribute sets during transmission, the framework leverages blockchain technology, using zero-knowledge proofs and Merkle verification to verify the authenticity of users’ identities and their attribute sets. Additionally, a new data structure, the Bitmap Attribute Merkle Tree (BAMP), is introduced to optimize the storage and retrieval of attribute information. This improves the system’s efficiency in large-scale user scenarios, providing a new solution for dynamic updates and access control of attribute information. The experimental results based on this framework demonstrate that the proposed solution achieves fine-grained access control in the data-sharing process while ensuring strong privacy protection and high efficiency in data retrieval.