Electricity data sharing not only enhances the scientific and precise decision-making process, but also serves as a crucial driving force for the intelligent upgrading of smart grid. However, if the sensitive information embedded within electricity data is leaked during the sharing process, it will pose serious security risks and threats. Currently, attribute-based encryption, as one of the important techniques for achieving secure data sharing, is widely used. However, its assumption of sender honesty is particularly fragile in complex network environments. In response, Susilo et al. (TDSC’21) proposed a sanitizable access control system by randomizing incoming ciphertexts to resist malicious senders deviating from the encryption algorithm, but this method still needs to be optimized in terms of efficiency and security assumptions. To overcome these limitations, this study design a privacy-preserving data sharing mechanism that significantly improves the scalability and operational efficiency while retaining the advantages of the original solution. Then, we reduce the security of the scheme to a more efficient standard assumption. Finally, performance analysis validates its efficiency, which provides a lightweight solution for resisting malicious senders in smart grid data sharing.