The answer, up front
Medical device makers and health systems face a migration shaped by two facts: regulators are setting 2030 and 2035 milestones, and patient records need decades of protection. That combination makes harvest-now-decrypt-later an active concern, not a future one. The current reporting leans on vendor-authored pieces and a preprint, so the claims below should be read as viewpoints and estimates rather than settled fact, but the direction is consistent with the government timelines that anchor it.
Devices and deadlines
Two Medcrypt authors wrote in October 2025 that medical device encryption could be broken within a few years to a decade and that regulators are setting deadlines 1. They say NIST will deprecate RSA and ECC by 2030 and phase out legacy algorithms by 2035, that the EU expects high-risk systems to reach post-quantum cryptography by the end of 2030 and medium and low risk by 2035, and that the FDA's June 2025 guidance asks for current NIST-recommended cryptography without naming a post-quantum approach 1. They advise crypto agility and a cryptography bill of materials 1. Medcrypt sells device cybersecurity, so these are vendor claims.
The UK clock applied to health
In the United Kingdom, the National Cyber Security Centre set three milestones in March 2025: complete discovery of cryptography-dependent systems by 2028, finish high-priority migration by 2031, and complete full migration by 2035 2. A July 2025 post from an IT services vendor applies that clock to the NHS, treating harvest-now-decrypt-later as an active threat to stolen patient data, noting that patient records need decades of protection, and naming legacy trust middleware as the main readiness obstacle 3. That post is vendor marketing, so weigh its framing accordingly, but the NCSC dates behind it are primary.
Device constraints
A December 2024 arXiv preprint on healthcare IoT says pacemakers and glucose monitors run on roughly 32 to 64 KB of RAM and 16 to 32 MHz processors, which complicates post-quantum algorithms, and it discusses interception and device-manipulation risks alongside harvest-now exposure of patient data 4. As read, the preprint is not peer reviewed, so its specifics are claims rather than confirmed findings, though the constraint it describes is real and widely acknowledged.
What this means for healthcare organizations
Retention is the planning fact. A record created today may need to stay confidential for a lifetime, which is exactly what makes harvest-now exposure present rather than future. Health systems and device makers should inventory where vulnerable cryptography protects long-lived records and devices, prioritize by sensitivity and retention, and build crypto agility and a cryptography bill of materials into device design and procurement.
How ArcQubit approaches it
ArcQubit treats the cryptographic inventory as the first deliverable. QuCode discovers where vulnerable cryptography sits across records, devices and cloud platforms, and QuTrust holds the migration plan so a health system can show a current, defensible posture as FDA, NCSC and EU timelines bite.