Study: AI Could Speed Quantum Threats — And Help Build Defenses

AI and Quantum Threats: A Dual‑Edged Sword

A recent analysis in the Journal of Quantum Information Science argues that artificial intelligence is set to become a catalyst for both the acceleration of quantum attacks on today’s encryption and the development of practical defenses. The paper, authored by Troy Coienth Troublefield of Capitol Technology University, frames AI and quantum computing as a single, intertwined cybersecurity challenge. Below we unpack the key take‑aways and outline actionable steps for security teams.

1. How AI Might Speed Up Quantum Attacks

• Improved error‑correction: Quantum computers are still hampered by noise and decoherence. Machine‑learning‑driven decoders can interpret error‑checking data more accurately, potentially narrowing the gap between experimental devices and the fault‑tolerant machines needed for Shor’s algorithm.
• Optimized algorithms and hardware estimates: Recent research (e.g., Craig Gidney’s 2025 estimate) suggests that breaking RSA‑2048 could require fewer than one million physical qubits—far less than earlier projections. While AI is not the sole driver of this reduction, its role in refining algorithms and simulation tools contributes to tighter resource forecasts.
• Harvest‑now, decrypt‑later: Adversaries could collect encrypted traffic today and store it for future decryption. AI can help sort massive data sets, flagging high‑value records (medical, intellectual‑property, classified) that are likely to retain sensitivity for decades.

2. Immediate Defensive Uses of AI

• Automated cryptographic discovery: Large enterprises often lack a clear inventory of where public‑key cryptography (RSA, ECC) is used. AI‑enabled scanners can locate cryptographic identifiers in code, binaries, and network flows, and even infer usage patterns where explicit markers are absent.
• Prioritization and tracking: Once vulnerable assets are mapped, machine‑learning models can rank them based on the sensitivity of the data they protect and the difficulty of migration, helping teams focus on the most critical replacements first.
• Cryptographic agility: AI tools can monitor software supply chains for new dependencies and alert administrators when a component introduces a deprecated algorithm, supporting a continuous‑update posture.

3. The Emerging Landscape of Quantum‑Resistant Cryptography

The National Institute of Standards and Technology (NIST) released its first three post‑quantum cryptography (PQC) standards in August 2024, covering key‑exchange and digital‑signature schemes that run on conventional hardware. The challenge now lies in deployment:

1. Identify long‑lived data – medical records, patents, and classified documents often require protection for decades.
2. Maintain an up‑to‑date inventory – AI‑driven discovery tools keep this list current as systems evolve.
3. Plan backward from deadlines – Use the inventory to set realistic migration timelines, accounting for dependencies and testing cycles.
4. Include AI assets – Model files, training data, and inference pipelines should be treated like any other high‑value secret.

4. Quantum‑Assisted Cyber‑Defense: Still Mostly Theory

Research exploring quantum computers for intrusion detection shows promising simulation results, but real‑world deployments are limited. Barriers include:

• Hardware error rates that can corrupt quantum‑based models.
• Interpretability – Security analysts need clear explanations for alerts, which quantum models currently lack.
• Cost and scalability – Quantum processors remain expensive and specialized.

Consequently, while quantum‑enhanced AI may become a future tool, organizations should focus on proven, near‑term measures.

Practical Takeaways

• Start with a cryptographic inventory using AI‑enabled discovery tools; verify findings manually to avoid false positives.
• Prioritize data with long confidentiality requirements and map its cryptographic dependencies.
• Adopt NIST‑approved PQC algorithms where feasible, and design systems for easy algorithm swaps (cryptographic agility).
• Monitor AI‑related assets for long‑term secrecy needs and include them in the migration plan.
• Stay informed about quantum hardware progress and emerging AI‑quantum research, but allocate resources to actions that deliver immediate risk reduction.

By treating AI and quantum computing as a combined threat vector, security teams can turn a potential accelerator of attacks into a powerful ally for proactive defense.