CAREER: Cryptographic Proofs, Outside the Black-Box

Cryptography has advanced far beyond its original goal of securing communication and today, allows us to compute on data in provably secure ways. Users regularly depend on untrusted service providers to perform an array of services like statistical analysis and learning over outsourced, distributed private data. In these settings, it is important to ensure security against adversaries that behave arbitrarily and deviate from protocol specifications. Cryptographic proofs form the bedrock of security against such adversaries, and have facilitated a multitude of powerful capabilities. This project tackles fundamental goals in the theory of cryptographic proofs: Can we efficiently verify the results of inefficient computation? Can we write proofs of statements that hide all secrets about them? Despite decades of research, there are crucial gaps in our understanding of these problems. Existing solutions either rely on external sources of trust or point to unknowns in our understanding due to heuristic security analyses in idealized models. This project overcomes these issues by improving how we reason about security. This involves investigating new reasoning techniques that meaningfully relax definitions and use the code of an adversary against the system. The project also builds a research community through workshops as well as training graduate, undergraduate and K-12 students in the foundations of cryptography and security. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.