VI - Tears for Quantum Fears

## Overview

Dr. Mark Carney, a technical specialist from Bank of England, delivers a comprehensive examination of quantum computing's intersection with cybersecurity. Rather than relying on hype ("the wind is coming"), Carney emphasizes facts, figures, and critical analysis. The talk progresses from quantum fundamentals through security implications to practical applications.

## Key Topics

### Quantum Fundamentals Carney explains that quantum effects—specifically superposition and entanglement—form the basis of quantum computing. A qubit differs from classical bits by existing in multiple states simultaneously, visualized via the Bloch sphere model. Superposition allows qubits to occupy a continuum of states between 0 and 1, while entanglement links particles such that measuring one reveals information about others. Quantum circuits combine quantum gates (Pauli X, Hadamard, controlled NOT) as 2×2 matrices operating on 2-element vectors.

### Quantum Threats to Cryptography Two algorithms pose theoretical threats: Shor's algorithm factors large numbers exponentially faster than classical methods, threatening RSA and ECC; Grover's algorithm achieves quadratic speedup in symmetric cipher attacks, effectively halving key lengths (AES-256 becomes AES-128 in security). However, timelines remain uncertain: Carney estimates 7–8 years minimum for these algorithms to become practical, assuming aggressive qubit scaling (doubling annually), but realistically 10–15 years accounting for engineering challenges. Current quantum systems have ~50–64 qubits; Shor's factoring requires millions.

### Post-Quantum Cryptography NIST's standardization effort (Round 3) aims to replace number-theoretic schemes with those resistant to quantum attacks. The leading approach is lattice-based cryptography, built on the Learning with Errors (LWE) problem: given a random array, secret, and noisy output, recovering the secret is NP-hard even for quantum computers. Finalist candidates include Kyber, Dilithium, NTRU, FrodoKEM, and SABER. The Carney notes NIST will likely standardize multiple algorithms rather than a single cipher, with complex requirements and key-length variations.

### Quantum Key Distribution BB-84 (1984, Bennett & Brassard) and E-91 protocols use quantum properties for provably secure key exchange. Alice encodes bits via photon polarization in one of two bases; Bob measures using a randomly chosen basis. They publicly compare bases and discard mismatches; eavesdropping increases quantum bit error rates, revealing Eve's presence. However, QKD has practical limitations: it requires dedicated fiber links (cannot packet-switch photons), fails in cloudy conditions, and suffers from atmospheric dispersion. Historical attacks include the 2011 laser-brightness attack revealing measurement phases, and acoustic side-channels from optical switching. Vendors: ID Quantique, MagiQ, Toshiba.

### Quantum Random Number Generation Carney demonstrates a hybrid CSPRNG combining local entropy with high-quality quantum randomness from IBM Q's 15-qubit systems. Hadamard gates create superposition with equal 0/1 probability. The practical implementation uses Blum Blum Shub CSPRNG, hashes together local and quantum sources, and scales to billions of bits. This proof-of-concept (unsuitable for production) shows quantum computing's near-term utility beyond cryptanalysis.

## Technical Details

- **Quantum timeline assumptions**: Doubling qubits annually; qubit coherence improving from microseconds to days/weeks - **Post-quantum hardness**: Lattice shortest vector problem has no known quantum algorithmic advantage - **Symmetric cipher defense**: Use AES-256 in GCM mode; Grover's requires exponentially more qubits per key-length increase - **RSA defense**: Migrate to 4096-bit keys or P-384 ECC; these are interim measures - **QKD error term (Eve detection)**: Rising bit error rates during key sifting indicate eavesdropping - **Quantum algorithm resources**: Qiskit (IBM's Python framework), Quantum Algorithm Zoo, IBM Quantum Experience (free cloud access)

## Takeaways

1. **Timeline realism**: Quantum threats are 7–15 years away under optimistic assumptions; complacency is unwarranted but panic is premature. 2. **Short-term mitigations**: Longer keys (4096-bit RSA, AES-256), AES-GCM mode, and preparation for NIST standards now. 3. **Post-quantum readiness**: Organizations should monitor NIST standardization; expect multiple algorithms with different trade-offs, not one "AES replacement." 4. **QKD limitations**: Useful for high-value point-to-point links but impractical for general cloud infrastructure and vulnerable to side-channel attacks. 5. **Quantum computing opportunities**: Beyond cryptanalysis, quantum systems enable financial modeling, molecular simulation, and cryptographically secure randomness generation. 6. **Knowledge investment**: Learning Qiskit and quantum gates now is prudent; high school linear algebra suffices for foundations.