Crypto Dark Matter

Crypto Dark Matter refers to a family of candidate PRF constructions proposed by Boneh, Ishai, Passelègue, Sahai, and Wu — BIP+18. The constructions mix operations in different algebraic domains — specifically, linear operations over (XOR) and over (mod-3 addition) — to build PRFs that are conjectured to be secure under new “mixed moduli” assumptions not known to be equivalent to standard assumptions like LWE or LPN.

The name reflects the idea that large regions of the cryptographic assumption landscape are unexplored (“dark”), and these constructions explore that space.

Assumption

The core hardness assumption concerns the difficulty of distinguishing the output of a “mod-2/mod-3” computation from random. Let be defined by alternating linear layers over and applied to the input bits (treating bits as elements of both fields simultaneously). The assumption is that no efficient adversary can distinguish from uniform even given arbitrary queries.

Formal variants include:

  • -mixed moduli LWE: a variant of LWE where some operations are mod-2 and others are mod-3
  • Low-complexity PRF conjecture: the existence of a PRF computable in or even

Known Results

  • Low-complexity PRF candidates (in / ) based on mixed-moduli assumptions — BIP+18
  • Mixed-moduli PRFs → applications in MPC with low-communication preprocessing, leakage-resilient PRFs, and more — BIP+18
  • Low-complexity PRG in is required by the iO construction from well-founded assumptions — JLS21
  • The assumption is not known to follow from or imply standard lattice assumptions

Variations

Low-complexity PRFs (in / )

Separate from CDM, there is interest in PRFs computable by low-complexity circuits. A PRF in (constant depth, threshold gates) would have strong implications for MPC efficiency.

Pseudorandom correlation generators (PCG)

A related paradigm where short seeds expand to long correlated randomness useful in MPC; CDM-style assumptions can ground efficient PCG constructions.

Attacks

  • Ongoing cryptanalytic attention; several early CDM candidates have been partially broken or weakened
  • Algebraic attacks exploiting the mixed-moduli structure (Gröbner basis methods, linearization) remain the primary avenue
  • No full break of the core constructions in BIP+18 is known