Secure Computation

Secure computation, often expressed as secure multi-party computation SMPC, is a cryptographic paradigm that enables multiple parties to jointly compute a function over their private inputs without revealing them to one another. Core techniques include homomorphic [encryption](/en/terms/homomorphic-encryption), secret sharing, and garbled circuits, each providing different trade-offs in communication, computation, and security guarantees. SMPC protocols are defined under adversary models such as semi-honest or malicious and can be instantiated for two-party or multiparty settings. Practical deployments consider performance overheads such as communication complexity and latency, and rely on careful protocol design to protect inputs, outputs, and intermediate states. Real world use cases span finance, healthcare, and machine learning, where privacy-preserving data analysis is essential. Limitations include scalability challenges, integration with existing systems, and the need for robust threat modeling. The field continues to evolve with federated learning, secure enclaves, and hybrid approaches that mix cryptography with trusted hardware.