Key Takeaways:
- Starknet published a Memory Protocol Draft on July 14 for user-owned AI data.
- The framework proposes onchain storage and management of personal AI data.
- The draft could position Starknet as infrastructure for AI data sovereignty.
Back
Starknet's July 14 memory draft targets user-owned AI data onchain
Starknet published a Memory Protocol Draft on July 14 proposing a framework for user-owned AI data storage and management on its blockchain infrastructure.
The draft, released by the Starknet core development team, outlines a protocol for storing personal AI data onchain, giving users cryptographic control over information generated through artificial intelligence applications, according to the proposal.
The framework addresses a structural tension in the AI economy: user data generated through AI interactions is typically controlled by centralized platforms. Starknet's proposal would let users store that data on its blockchain, with permissioning enforced by smart contracts rather than platform terms of service.
If adopted, the Memory Protocol could position Starknet as a leading infrastructure layer for AI data sovereignty, attracting developers building at the intersection of AI and crypto. The draft is open for community feedback, with implementation timelines subject to Starknet's governance process.
The proposal arrives as the broader crypto industry seeks to establish a role in the AI data economy, where user-generated training data and interaction histories have become valuable assets controlled by a small number of technology companies. Starknet, an Ethereum layer-2 scaling solution using zero-knowledge rollup technology, serves as a platform for decentralized applications.
The Memory Protocol Draft specifies how AI-related data — including user preferences, interaction histories, and model fine-tuning data — could be stored and managed on Starknet's blockchain. The framework emphasizes user ownership, meaning individuals would retain cryptographic control over their data rather than surrendering it to AI application providers.
This approach aligns with a broader push within the crypto industry toward decentralized identity and data sovereignty, areas that have gained urgency as AI adoption accelerates. Projects building on Starknet could use the Memory Protocol as a foundation for AI applications that respect user data rights by design.
Competing layer-1 and layer-2 networks, including those on Ethereum and Solana, are also developing frameworks for user-owned data infrastructure. Starknet's focus on zero-knowledge proofs could provide privacy advantages for AI data management, as the technology allows verification without revealing underlying data.
The draft is subject to community review and governance processes on Starknet. No specific timeline for implementation has been announced.
This article is for informational purposes only and does not constitute investment advice.
[1] Starknet Memory Protocol Draft Puts User-Owned AI Data On The Crypto Agenda[2] Binance Marks Nine Years: Nearly Half of All Crypto Holders Use Binance, $156 Trillion in All-Time Volume, and a Growing Vision Beyond Crypto[3] SonicStrategy Signs LOI to Acquire AI Infrastructure Assets from ARKLAB AI[4] TxFlow L1 launches Probly, expanding its multi-application ecosystem[5] Coinsilium Group Limited: Predictive Labs Unveils Nijinn; Coinsilium Increases Strategic Investment and Accelerates Venture-Building Strategy
