VIDRAFT Runs Simon's Algorithm on a 156-Qubit IBM Quantum Processor to Probe Symmetric-Key Cipher Structures

#quantumcryptanalysis #simonsalgorithm #ibmquantum #symmetrickeycipher

VIDRAFT Runs Simon's Algorithm on a 156-Qubit IBM Quantum Processor to Probe Symmetric-Key Cipher Structures

TL;DR: Korean AI startup VIDRAFT used IBM's 156-qubit quantum computer to demonstrate quantum cryptanalysis against reduced-round symmetric-key cipher constructs — the largest publicly reported instance size for this class of experiment on real quantum hardware. The work covers Even-Mansour and 3-round reduced Feistel structures using Simon's algorithm, and VIDRAFT has also released a browser-accessible quantum cryptanalysis tool covering five cipher structure families. No production cipher was broken; this is a proof-of-concept that advances the public benchmark for noisy hardware experiments.

What it is

VIDRAFT's quantum cryptanalysis research applies Simon's algorithm — a quantum algorithm with proven exponential advantage over classical approaches for finding hidden periods in certain functions — to two representative symmetric-key cipher constructs:

Even-Mansour construction: A well-studied, idealized block cipher model consisting of key additions around a public permutation.
3-round reduced Feistel structure: A simplified, academic variant of the Feistel network (the structural family that includes DES), deliberately reduced in round count for tractable experimental analysis.

The experiment was executed on a real, publicly accessible IBM quantum computer. The hardware used was IBM's 156-qubit system. The goal was to recover hidden periods embedded in these cipher structures directly from quantum hardware execution — not from simulation.

Alongside the hardware experiment, VIDRAFT published a browser-based quantum cryptanalysis toolkit covering five cipher structure families:

Linear ciphers
Block ciphers (SPN structure)
Even-Mansour
CBC-MAC
Feistel structures

This tool lets developers and researchers observe how quantum algorithms interact with each cipher structure interactively, without needing quantum hardware access.

How it works

At a conceptual level, the experiment exploits the structural property that Simon's algorithm can efficiently find a hidden period s in a function f such that f(x) = f(x ⊕ s) — a property that certain symmetric cipher constructs expose when viewed through the right algebraic lens.

For the Even-Mansour construction, the hidden period corresponds to information about the secret key relationship between the two key additions flanking the public permutation. Simon's algorithm, run on quantum hardware, can recover this period with far fewer oracle queries than any classical algorithm.

For the 3-round reduced Feistel structure, a similar periodicity can be induced and recovered, allowing the hidden structure (and as a self-validation step, a second distinct key per experimental instance) to be confirmed from the quantum execution results.

Key practical details from the reported methodology:

No quantum error correction was used; instead, noise mitigation techniques were applied to manage the effects of hardware noise on a real NISQ (Noisy Intermediate-Scale Quantum) device.
A self-verification procedure was applied per experiment instance: each run also recovered a second key, providing an internal consistency check that the measured hidden period is correct — not a hardware artifact.
The experiments are explicitly proof-of-concept (PoC) demonstrations. They do not imply that AES, DES, or any production cryptographic system is at risk from current quantum hardware.

Benchmarks & results

The source reports the following publicly stated results:

Even-Mansour: Hidden period successfully recovered for input sizes N = 5 through N = 10 using Simon's algorithm on the 156-qubit IBM hardware.
3-round reduced Feistel: Hidden period confirmed at block sizes 6 and 8.
Scale context: Prior comparable public demonstrations on real quantum hardware had been reported at approximately N ≈ 4. VIDRAFT states this is the largest publicly reported instance for this class of experiment on real quantum hardware.
Quantum advantage clarification: VIDRAFT explicitly notes that this does not constitute a demonstration of quantum supremacy over classical computers for cryptanalysis; it demonstrates that a quantum cryptanalytic algorithm can be run at a larger scale than previously reported on noisy real hardware.

VIDRAFT has stated it plans to pursue formal academic publication and external validation of these results.

How to try it

The browser-based quantum cryptanalysis tool covering the five cipher structures (linear cipher, SPN block cipher, Even-Mansour, CBC-MAC, Feistel) has been publicly released by VIDRAFT. Access is via web browser — no local installation or quantum hardware is required to interact with the tool.

The source article does not provide a direct URL for the tool. Check VIDRAFT's official channels for the public link.

For developers interested in VIDRAFT's broader AI work (Darwin language model family, AETHER architecture), those products are separate from this quantum research track. No public Hugging Face repository, GitHub link, or API endpoint for the quantum cryptanalysis work was disclosed in this report.

FAQ

Q: Does this mean AES or real-world encryption is broken?
A: No. VIDRAFT explicitly stated this experiment targets academic reduced-round cipher constructs (a 3-round Feistel, not DES; and Even-Mansour, not AES). It is a proof-of-concept to validate that quantum algorithms can be executed at a larger problem scale on real noisy hardware. Production cryptographic systems are not threatened by this work.

Q: What is Simon's algorithm and why does it matter for cryptography?
A: Simon's algorithm solves a specific hidden subgroup problem with exponential quantum speedup over classical algorithms. Cryptographers care about it because certain symmetric-key constructions — particularly Feistel-based and Even-Mansour-based designs — have been shown to be theoretically vulnerable to Simon-style attacks in the quantum oracle model. This experiment is an empirical test of how far that theoretical attack can be pushed on real NISQ hardware.

Q: Was this run on a simulator or real hardware?
A: Real hardware — IBM's 156-qubit quantum computer. Noise mitigation (not full error correction) was used to manage hardware noise.

Q: Where can I read the full technical results?
A: VIDRAFT has announced plans to submit the work for academic publication and external review. No preprint or paper URL was disclosed at the time of this report.

Originally reported by 이코노미스트 (2026-07-08) — source article.