Secure Multiparty Computation

MPC allows multiple parties to compute a function over their collective inputs while keeping those individual data points private.

Secure Multiparty Computation (MPC) eliminates the need for a trusted third party by distributing a computation across several nodes. Each participant holds a cryptographic share of the data (e.g., using Shamir's Secret Sharing) so that no single entity ever sees the full dataset. This technology powers high-stakes applications like the Boston Women’s Workforce Council's wage gap analysis and Google’s private ad conversion tracking. By ensuring that only the final output is revealed (the sum, the average, or the intersection), MPC protects sensitive assets like cryptographic keys and genomic data from internal leaks and external breaches.

https://mpcalliance.org