Indistinguishability Obfuscation from Well-Founded Assumptions

Indistinguishability obfuscation, introduced by [Barak et. al. Crypto 2001], aims to compile programs into unintelligible ones while preserving functionality. It is a fascinating and powerful object that has been shown to enable a host of new cryptographic goals and beyond. However, constructions of indistinguishability obfuscation have remained elusive, with all other proposals relying on heuristics or newly conjectured hardness assumptions. In this work, we show how to construct indistinguishability obfuscation from subexponential hardness of four well-founded assumptions. We prove: Informal Theorem: Let. tau is an element of(0, infinity), delta is an element of(0, 1), epsilon is an element of(0, 1) be arbitrary constants. Assume sub-exponential security of the following assumptions: - the Learning With Errors (LWE) assumption with subexponential modulus-to-noise ratio 2(k epsilon) and noises of magnitude polynomial in k, where k is the dimension of the LWE secret, - the Learning Parity with Noise (LPN) assumption over general prime fields Z(p) with polynomially many LPN samples and error rate 1/l(delta), where l is the dimension of the LPN secret, - the existence of a Boolean Pseudo-Random Generator (PRG) in NC0 with stretch n(1+tau), where n is the length of the PRG seed, - the Decision Linear (DLIN) assumption on symmetric bilinear groups of prime order. Then, (subexponentially secure) indistinguishability obfuscation for all polynomial-size circuits exists. Further, assuming only polynomial security of the aforementioned assumptions, there exists collusion resistant public-key functional encryption for all polynomial-size circuits.