TL;DR: Quantum computers solve certain math problems exponentially faster than classical machines, and those exact problems—factoring large integers and computing discrete logarithms—underpin RSA, ECC, and Diffie-Hellman encryption. Once a sufficiently large, error-corrected quantum computer exists, today’s public-key infrastructure collapses, forcing a migration to post-quantum algorithms.
The Math That Undoes Modern Security
RSA’s security rests on the difficulty of factoring a 2048-bit number into its prime components. A classical supercomputer would need billions of years. Shor’s algorithm, running on a fault-tolerant quantum computer, reduces that to hours or days by exploiting superposition and entanglement to evaluate vast numbers of possibilities simultaneously. The same algorithm breaks elliptic-curve cryptography, which protects most TLS traffic today, using far fewer qubits than RSA requires.
If you want to dig deeper, check out our guide on Quantum-Safe Encryption Goes Mainstream: What You Need to Kn.
Where the Hardware Actually Stands
Current systems remain far from the threshold. IBM’s Condor processor holds 1,121 superconducting qubits; Atom Computing has demonstrated over 1,000 neutral-atom qubits. But raw qubit counts mislead. Breaking RSA-2048 is estimated to require roughly 20 million noisy physical qubits, or about 4,000 logical, error-corrected ones. Today’s machines produce logical qubits in the dozens at best, with error rates around 10⁻³. Google’s Willow chip, announced in late 2024, crossed a milestone by showing that adding qubits reduced errors—evidence that error correction can scale. IBM’s roadmap targets 100,000 qubits by 2033 and full fault tolerance beyond that.
Why “Harvest Now, Decrypt Later” Matters
Adversaries are already collecting encrypted data today, betting they can decrypt it once quantum hardware matures. For data with long confidentiality requirements—medical records, state secrets, financial archives—the threat is active now, not in 2035.
Industry Response
NIST finalized its first post-quantum standards in August 2024: ML-KEM (CRYSTALS-Kyber) for key encapsulation and ML-DSA (CRYSTALS-Dilithium) for signatures, with FIPS 203, 204, and 205 published. Cloud providers, browser vendors, and messaging apps have begun hybrid deployments that combine classical and post-quantum key exchange. The transition will take a decade or more, which is precisely why standards bodies and enterprises are starting now rather than waiting for the hardware to arrive.
FAQ
Q: When will quantum computers actually break RSA?
A: No credible expert expects it before the 2030s, and possibly much later. The barrier is error correction, not qubit count, so timelines depend on engineering breakthroughs that haven’t happened yet.
Q: Should I stop using RSA today?
A: Not urgently for casual browsing, but organizations handling long-lived sensitive data should begin migrating to NIST’s post-quantum standards now, since encrypted traffic captured today could be decrypted later.
Q: Are post-quantum algorithms unbreakable?
A: No. They rely on different hard problems—lattice-based cryptography, for instance—that resist known quantum attacks. They could still fall to future classical or quantum cryptanalysis, which is why hybrid approaches are recommended.
Leave a Reply