Quantum Computing: A 2025 Reality Check

Quantum computing attracts more hype per useful result than perhaps any other technology field. The press release announcing "quantum supremacy" in 2019 (Google's Sycamore processor completing a benchmark in 200 seconds that would take a classical supercomputer 10,000 years) was followed by IBM demonstrating the same task could be done classically in 2.5 days — then in seconds with better algorithms. The field is real and progressing, but calibration matters.

What Quantum Computers Are

Classical computers process information as bits — 0 or 1. Quantum computers use qubits, which exploit quantum mechanical superposition to exist in combinations of 0 and 1 simultaneously. Quantum entanglement allows qubits to be correlated in ways that enable certain computations to be performed exponentially faster than any known classical algorithm.

The key word is "certain." Quantum speedups apply to specific problem classes: factoring large integers (Shor's algorithm), searching unsorted databases (Grover's algorithm), simulating quantum systems, and optimization problems with specific mathematical structure. They do not provide general speedups over classical computers.

The Error Problem

Current quantum computers are noisy intermediate-scale quantum (NISQ) devices. The qubits in these systems are highly sensitive to environmental disturbance (decoherence), and every operation introduces errors. As of 2025, the best systems have error rates of roughly 0.1-1% per gate operation. Running Shor's algorithm to break RSA-2048 would require millions of logical qubits — each logical qubit requiring hundreds to thousands of physical qubits for error correction. Current systems have hundreds to low thousands of physical qubits.

Google's 2024 demonstration with its Willow chip showed meaningful progress in error correction: as the system scaled, errors decreased rather than accumulated — a necessary (if not sufficient) condition for fault-tolerant quantum computing. This is genuinely significant. But it is a proof of principle, not a deployable system.

Where Quantum Actually Helps Now

The near-term commercial case for quantum computing is narrower than headlines suggest. Quantum simulation of molecular systems is the most credible near-term application — pharmaceutical companies are investing in quantum chemistry for drug discovery because even small quantum systems can simulate molecular interactions that are classically intractable. Optimization problems in logistics and finance may yield quantum advantage at scales achievable in the next 5-10 years.

The Cryptography Timeline

The threat to current public-key cryptography (RSA, ECC) from quantum computers is real but not imminent. NIST finalized its first post-quantum cryptographic standards in 2024. Organizations handling data that must remain secure for 10+ years should begin migrating. For most practical purposes, RSA-2048 is not at immediate risk — the timeline to cryptographically relevant quantum computers is likely 10-20 years, not 2-5.

What to Watch

The genuine milestones to watch are: demonstration of a logical qubit with error rates below the fault-tolerance threshold, execution of a quantum algorithm on a practically useful problem that cannot be matched by classical simulation, and scaling of logical qubit counts into the hundreds. None of these have been demonstrated as of early 2025.