Quantum Computing and the Future of Encryption: What You Need to Know
Quantum computers will eventually break today's encryption. The good news is getting started to become PQC resistant is actually not as hard as you think.
The Plot of a Movie That's About to Become Real
Recently, I rewatched "Now You See Me 2." Back in 2016, the movie's plot — where the bad guy gets hold of a chip that can read data on any phone and computer — felt like pure sci-fi fantasy. But now in 2026, with all the post-quantum computing developments, I realize this isn't fantasy at all. It could happen soon.
So I wrote this post as a starting point for everyone to start paying attention to post-quantum computing and what it means for the security of our modern world.
What is Quantum Computing?
To understand the threat, it helps to understand how computing works today and what's changing.
Classical Computing: The Light Switch
Every classical computer — your laptop, your phone, the servers running your applications — processes information using bits. A bit is either 0 or 1. Think of it as a light switch: on or off. Every calculation, every video you stream, every transaction you process is ultimately a sequence of these on/off switches flipping very fast.
Classical computers are incredibly powerful for most tasks. But certain mathematical problems — like finding the prime factors of a very large number — are so computationally expensive that even the fastest supercomputers would take longer than the age of the universe to solve them. This mathematical difficulty is exactly what modern encryption relies on.
Quantum Computing: The Dimmer Switch
Quantum computers work differently. Instead of bits, they use qubits. A qubit isn't limited to 0 or 1 — think of it as a dimmer switch that can be anywhere between fully off and fully on, and every position in between, all at the same time.
This is possible because of a property called superposition: a qubit can represent multiple states simultaneously. Combined with two other quantum properties — entanglement (qubits can be linked so that the state of one instantly affects the other) and interference (quantum algorithms can amplify correct answers and cancel wrong ones) — quantum computers can explore many possible solutions in parallel.
For most everyday tasks, this doesn't matter. Classical computers handle them just fine. But for specific mathematical problems — particularly the ones that protect our encryption — quantum computers offer a fundamentally different, and much faster, approach.
Where Are We Today?
Quantum computing is real, but still in its early stage — known as the NISQ (Noisy Intermediate-Scale Quantum) era. Current quantum processors range from tens to over a thousand physical qubits, depending on the hardware approach [1]. But these are noisy physical qubits prone to errors. To break modern encryption, we'd need thousands of error-corrected logical qubits — which requires millions of physical qubits with today's technology.
The science is proven. The engineering to build a machine large and stable enough is the remaining challenge. Most experts estimate this is 10 to 15 years away. It's not a question of "if" — it's a question of "when."
What Does Quantum Computing Mean for Cryptography?
Here's where it gets real. The encryption protecting most of the internet today — every time you see the padlock icon in your browser, every API call your application makes, every VPN tunnel — relies on asymmetric cryptography. And asymmetric cryptography is exactly what quantum computers can break.
The Core Problem
Asymmetric encryption (RSA, ECC, Diffie-Hellman) works by generating a pair of keys: a public key and a private key. The security depends on the fact that deriving the private key from the public key requires solving a math problem that classical computers can't solve in any reasonable time.
In 1994, mathematician Peter Shor published an algorithm proving that a quantum computer could solve these exact math problems efficiently. It doesn't brute-force the answer — it takes a mathematical shortcut that classical computers simply can't take.
Once a sufficiently powerful quantum computer exists, an attacker could take any public key, run Shor's algorithm, and derive the private key. With the private key, they can decrypt communications and forge digital signatures.
What About AES?
Not all encryption is at risk. Symmetric encryption like AES-256 — used to encrypt the actual data in most communications — has no mathematical structure for quantum algorithms to exploit. The only attack is brute-force: try every possible key. A quantum algorithm called Grover's can speed this up, but even with that speedup, breaking AES-256 would take roughly 10^19 years. Your data encrypted with AES-256 is safe.
The catch: however, AES keys are typically exchanged using asymmetric cryptography. So while the lock on your data is strong, the method used to hand over the key to that lock is vulnerable.
"Harvest Now, Decrypt Later"
This is the threat that makes PQC urgent today, not in 10 years.
Nation-state actors and sophisticated adversaries are already recording encrypted internet traffic. They can't read it now. But they're storing it. Once a cryptographically relevant quantum computer exists, they can go back, break the key exchange from the captured traffic, recover the AES session keys, and decrypt everything.
For data that needs to stay confidential for 10 to 25+ years — government communications, healthcare records, financial data, intellectual property, trade secrets — this is a real and present risk. The data is being harvested today. The decryption will happen later.
What Exactly Breaks?
What | Crypto Used | Quantum Threat |
|---|---|---|
Key exchange (TLS handshake) | ECDH / RSA | Shor's algorithm breaks it — attacker gets the session key |
Digital signatures (certificates) | RSA / ECDSA | Shor's breaks it — attacker can forge identities |
Data encryption | AES-256 | Grover's algorithm — still safe with 256-bit keys |
The bottom line: the encryption on your data is fine. The way you exchange keys and verify identities is not.
The good news: to build quantum-resistant systems, we don't need quantum computers. The new post-quantum cryptographic algorithms run on the same classical hardware we use today. It's a software upgrade, not a hardware replacement.
What is AWS Doing?
AWS has been working on post-quantum cryptography (PQC) for over five years. AWS employees contributed to the NIST PQC standards published in August 2024, and AWS is already shipping PQC capabilities in production services. For the full technical details, see the AWS Security Blog [2] and the AWS Post-Quantum Cryptography page [6]. Here's what everyone needs to know.
New Standards: The Replacement Algorithms
In August 2024, NIST published three new cryptographic standards designed to resist quantum attacks — FIPS 203 (ML-KEM), FIPS 204 (ML-DSA), and FIPS 205 (SLH-DSA) [3]. AWS is one of the contributors to these NIST post-quantum cryptography standards.
Protecting Data in Transit: PQ-TLS on AWS Endpoints
AWS is deploying hybrid post-quantum TLS (PQ-TLS) across all public service endpoints. "Hybrid" means the TLS handshake uses both classical ECDH and post-quantum ML-KEM together. An attacker would need to break both algorithms to compromise the connection.
Services already supporting PQ-TLS include AWS KMS, Secrets Manager, ACM, Payment Cryptography, Amazon S3, Elastic Load Balancing (ALB and NLB), API Gateway, CloudFront, and Transfer Family. AWS continues migrating all remaining HTTPS-based endpoints.
For most customers, the action is straightforward: upgrade your AWS SDK. Many SDKs enable PQ-TLS by default in recent versions. For example, the AWS SDK for Go enables PQ-TLS when you upgrade Go to v1.24+, and the AWS SDK for Java supports it in v2.35.11+ with AwsCrtHttpClient. Other SDKs that rely on system OpenSSL (Python, PHP, Ruby, CLI) need OpenSSL 3.5+. For the full SDK support matrix, see the Enabling Hybrid Post-Quantum TLS documentation [7].
Protecting Long-Lived Signatures: KMS and Private CA
For digital signatures that need to remain trusted for years or decades — firmware on IoT devices, code signing, device certificates — AWS has released PQC-resistant updates to two key services:
- AWS KMS now supports ML-DSA signing keys (launched June 2025) [4], protected in FIPS 140-3 Level 3 HSMs. Three key specs are available (ML_DSA_44, ML_DSA_65, ML_DSA_87) for different security levels.
- AWS Private CA now supports creating ML-DSA certificate authorities and issuing ML-DSA certificates (launched November 2025) [5] for internal use cases — mTLS between services, device authentication, and code signing with formal chain of trust.
Customers can start using these services today to issue quantum-resistant signatures and certificates for new workloads.
What's Still in Progress
Public-facing website certificates (the ones browsers validate) don't support PQC yet. This requires the entire ecosystem to align: CAs must issue ML-DSA certificates, browsers must validate them, and the CA Browser Forum must define new rules. AWS is actively participating in this effort through Amazon Trust Services.
Where to Start?
You don't need to fix everything at once. Start by understanding your exposure, then work through the highest-impact actions first.
Step 1: Know Where You Use Asymmetric Cryptography
Before you can act, you need to know what's at risk. Conduct a cryptographic inventory — identify every place your systems use asymmetric cryptography:
- TLS connections (HTTPS endpoints, API calls, service-to-service communication)
- Digital signatures (code signing, document signing, certificate issuance)
- Key exchange (VPN/IPsec tunnels, SSH connections)
- Certificate chains (mTLS, device authentication)
Step 2: Categorize by Risk
Not all asymmetric crypto usage is the same. Group your inventory based on category:
Category | Why It Matters |
|---|---|
Data in transit over the public internet | Vulnerable to "harvest now, decrypt later" today |
Signatures on long-lived artifacts (firmware, certs) | Must remain valid for years or decades |
Data in transit over private networks (VPC, PrivateLink) | Attacker needs internal network access to capture traffic |
Signatures on short-lived tokens (JWT, sessions) | Expires quickly, limited window of exposure |
Step 3: Follow the Decision Flow
Once you know where your asymmetric crypto is and how exposed it is, the next step depends on the use case:
The Key Takeaway
Post-quantum cryptography isn't a future problem. The "harvest now, decrypt later" threat means your data transmitted today could be decrypted in the future. The algorithms to defend against this are standardized, and AWS services already support them.
You don't need to understand quantum physics. You don't need quantum hardware. The transition to post-quantum cryptography is more about software and library upgrades than replacing hardware. For example, upgrading your AWS SDK to a version that supports PQ-TLS, adopting TLS 1.3, or using ML-DSA keys in AWS KMS for new signing workflows.
Build your inventory. Follow the decision flow. And keep an eye on AWS announcements as PQC support expands across more services.
References:
AI tools were used for editorial assistance in preparing this work.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article