Security and Privacy

This chapter introduces Nesa's cutting-edge hybrid approach to security and privacy enhancement. The essence of this hybrid design lies in the thoughtful integration of hardware-based and cryptographic-based solutions, each selected and optimized for varying scenarios within our ecosystem.

Our hybrid security and privacy is grounded in the recognition that privacy concerns manifest in different forms—users may wish to conceal their input data or the results of their inferences, while node owners might seek to protect the confidentiality of their model parameters. Our hybrid design acknowledges the unique requirements of these use cases by deploying the most appropriate privacy-preserving technologies.

Through the synergy of the robust, hardware-centric protections of Trusted Execution Environments (TEEs) and the advanced cryptographic techniques of zero-knowledge machine learning (ZKML), consensus-based distribution verification (CDV), and split learning (SL), we ensure that security and privacy are foundational pillars of the system.

This chapter elucidates the rationale behind Nesa's hybrid strategy, offering a comprehensive blueprint for achieving the highest standards of security and privacy while maintaining the usability and efficiency of the decentralized inference process.