PAI Data, Summary of the Project PAI Data Protocol

The Project PAI Data Protocol (“PAI Data”) is a specification that extends the Project PAI Blockchain Protocol to include a method of securing and provisioning access to arbitrary data. In the context of PAI Coin Development Proposal (PDP) 2, this paper defines two important transaction types that PAI Data supports: Storage Transactions, which facilitate storage of data and proof of ownership, and Sharing Transactions, designed to enable granting and revocation of data access to designated recipients. A comparative analysis of PAI Data against similar blockchain-based file storage systems is also presented.

💡 Research Summary

The paper presents the Project PAI Data Protocol (PAI Data), an extension of the Project PAI blockchain that adds native support for secure storage and controlled sharing of arbitrary data. The authors begin by outlining the limitations of conventional blockchains for data‑intensive applications: on‑chain storage is prohibitively expensive, and existing solutions rely on off‑chain services for both persistence and access control, which fragments trust and auditability. To address these gaps, PAI Data defines two novel transaction types.

1. Storage Transactions – A user encrypts the payload off‑chain, computes a cryptographic hash (SHA‑256), and embeds the hash together with optional hash of the encryption key into a PAI transaction. This on‑chain record serves as an immutable proof of ownership and integrity while the bulk data remains in an external storage layer (cloud, IPFS, or any distributed file system). By storing only metadata, the protocol respects blockchain size constraints yet guarantees non‑repudiation.
2. Sharing Transactions – The data owner generates a symmetric key for the encrypted payload, then re‑encrypts this key with each recipient’s public key using an ECIES‑style scheme. The resulting ciphertext is posted in a new PAI transaction, effectively granting the recipient the ability to decrypt the off‑chain data. Revocation is achieved by issuing a subsequent sharing transaction that either replaces the encrypted key with a null value or includes a time‑based expiry flag, making the previous key unusable.
   The protocol integrates the PAI token as an incentive mechanism. Storage providers receive token payments for hosting data, while validators earn rewards for confirming storage proofs and for verifying that sharing transactions correctly reference the intended recipients. This creates an economic loop that aligns the interests of data owners, custodians, and the broader PAI ecosystem.
   A comparative analysis follows, positioning PAI Data against prominent blockchain‑based storage projects such as IPFS, Filecoin, Storj, and Sia. IPFS offers content‑addressed addressing but lacks built‑in access control; Filecoin introduces storage contracts and replication guarantees but still depends on external applications for permission management. Storj and Sia provide encrypted, distributed storage with token economies, yet their permission models are not natively recorded on the underlying ledger. In contrast, PAI Data’s sharing transaction encapsulates both the grant and revocation of rights directly on the chain, delivering transparent audit trails and simplifying compliance. Moreover, because PAI Data leverages the existing PAI token, it can tap into an established user base without launching a separate economic layer.
   The authors acknowledge several challenges. First, the security of off‑chain storage remains a single point of failure; robust redundancy and reputation systems are required. Second, key management complexity grows with the number of recipients, potentially inflating transaction fees. Third, the on‑chain cost of frequent sharing transactions could hinder scalability, especially under high‑throughput scenarios. To mitigate these issues, the paper proposes future work on sharding the transaction pool, employing layer‑2 roll‑ups for batch sharing operations, and exploring post‑quantum cryptographic primitives to future‑proof the protocol.
   In conclusion, the Project PAI Data Protocol offers a cohesive framework that unifies data provenance, ownership proof, and fine‑grained access control within a single blockchain ledger. By coupling minimal on‑chain metadata with flexible off‑chain storage and a token‑driven incentive model, PAI Data achieves a balance of security, transparency, and economic viability that surpasses existing decentralized storage solutions. The work lays a solid foundation for further research into scalable, privacy‑preserving data ecosystems built on public blockchains.