Artificial intelligence (AI) is now widely adopted across fields, prompting AI companies to deploy models to the cloud for cost, resource, and scalability benefits. However, these cloud-hosted models face security and credibility challenges. Equipment failures or network attacks may compromise data integrity, while commercial interests might cause AI companies or cloud providers to deploy models deviating from their stated specifications, such as version or copyright details. To address these issues, we introduce AI-auditor, a novel data auditing framework that verifies both data integrity and model alignment with declared specifications. Using a challenge-response approach, AI-auditor maintains constant bandwidth usage and O(1) verification efficiency, regardless of file size. It also features a multikey management server mechanism to generate user keys, minimizing risks of single-point failures and trust issues. Analysis and simulations confirm AI-auditor’s correctness, security, and high execution efficiency.

As more internet users opt to store their data in cloud storage, ensuring data integrity becomes a paramount concern. The emerging provable data possession (PDP) scheme enables auditors to verify data integrity with reduced bandwidth consumption compared to hash-based alternatives. Nevertheless, most existing PDP variants rely on a centralized node for generating or maintaining user keys, creating a potential single point of failure. Moreover, previous PDP schemes could only detect whether challenged data blocks were corrupted, lacking the ability to pinpoint affected blocks precisely. To tackle these challenges, we propose a novel PDP scheme that eliminates the necessity for a key management center and supports the localization of corrupted data blocks. In our scheme, users no longer need to retain private keys once they cease performing data dynamic operations, thus liberating them from reliance on external entities for key maintenance. Moreover, the new scheme utilizes existing authenticators in the cloud to identify corrupted file blocks, eliminating the necessity of storing hash values for these data blocks as seen in most of existing implementations. This effectively reduces required storage space. Furthermore, we introduce SStore, a decentralized cloud storage platform that incorporates the new PDP scheme to verify data integrity. SStore facilitates public auditing of user data, thereby enhancing transparency in the data verification procedure. Moreover, SStore leverages basic algebraic operations for data auditing, significantly increasing its efficiency. We analyze the security of the new PDP scheme, and evaluate the performance of both the PDP scheme and SStore to demonstrate their efficiency.

The general data protection regulation (GDPR) is a European Union (EU) data protection and privacy law. According to the GDPR, the data on a hosting platform must meet semantic consistency and data integrity requirements. Semantic consistency means that the data operation should comply with the GDPR, while data integrity is meant to ensure that the outsourcing data should be intact. The two terms are not interchangeable. For example, if a cloud service provider migrates data to foreign storage nodes without authorization of the data owner, the data integrity requirement of the GDPR is met but the semantic consistency requirement is not. How to ensure data integrity and compliance is the main challenge for a GDPR-compliant data supervision platform. To achieve this aim, we leverage a blockchain-based data management framework to check the data compliance, which can break the black box of the data hosting platform and demonstrate its logic to data owners, allowing for inspection. We propose a new provable data possession (PDP) scheme for the aforementioned framework that can check for semantic consistency and data integrity simultaneously. The verifier does not need to hold any audited data, which can reduce bandwidth usage. The verification result can be regarded as the proof for subsequent data recovery and accountability. Experimental results show higher efficiency of the PDP scheme.

Emerging blockchain technology has become the cornerstone of many applications providing trusted data services. However, existing blockchain platforms cannot meet the growing demand for big data storage. Blockchain should duplicate both transactions and other user-defined data across nodes for integrity assurance. The rapid expansion of data on blockchain (on-chain data) increases the difficulty of deploying a full node, resulting in decreasing the degree of decentralization and adding the risk of broken data. To tackle these problems, we propose sChain, a novel framework for improving blockchain storage capacity, which does not revise blockchain implementation and can be applied to almost all the existing blockchain platforms. sChain outsources the user data to storage devices that are structurally external to the blockchain network. In theory, a user can outsource unlimited data to sChain. However, those off-chain data may suffer from corruption. To verify the data integrity, we propose a new provable data possession (PDP) scheme, which does not need a centralized entity to maintain any secret keys and therefore eliminates a single point of failure. What is more, we also design a prototype to accelerate the proposed PDP scheme through Intel SGX technology and parallel processing. Security analysis and evaluation results show that sChain can protect data security and effectively improve the blockchain storage capacity, respectively.