Medical device companies experience AI in two very different ways: as a regulated capability inside their products, and as an operational accelerant around them. The operational side is the quicker win; the in-product side follows a well-defined (and recently updated) FDA pathway. Here’s how, and how dgm implements it. (dgm implements osFoundry, a separate company’s platform — we are not osFoundry; device regulatory determinations stay with your regulatory team.)

What AI actually does for device companies

The honest framing: operational AI — R&D support, quality, documentation, regulatory writing — is the faster, lower-barrier win, while in-product AI (device software functions) follows the FDA device pathway. dgm focuses on the operational side and the right controls; your regulatory team owns device submissions.

High-value use cases

  • Operational R&D and documentation — accelerating design documentation and technical writing.
  • Quality and compliance support — easing quality-system documentation and review.
  • Regulatory writing assistance — drafting submission sections (regulatory-verified).
  • In-product AI (regulated) — AI-enabled device functions, governed by the FDA pathway below.

The regulatory reality: SaMD, PCCP, and QMSR

For anything touching the product, three things matter:

  • AI device functions are FDA-regulated as Software as a Medical Device.
  • PCCP (final guidance, December 2024). A Predetermined Change Control Plan lets you pre-specify, validate, and get authorization for future model changes at submission time — so qualifying updates can ship without a new submission, if intended use is unchanged. It also emphasizes bias mitigation and post-market surveillance.
  • QMSR (effective February 2026). The quality regime changed: revised 21 CFR Part 820 incorporating ISO 13485:2016 now governs design controls and computer-system validation — relevant to AI used in or around devices.

dgm builds operational/technical controls with these in mind; device regulatory determinations stay with your regulatory team.

How to start

Start with operational AI — documentation, quality, regulatory writing support — for a faster win, while your regulatory team manages any in-product AI through the SaMD/PCCP pathway under QMSR. Prove the operational value, then expand. dgm’s assessment finds the right starting point.

How dgm helps

dgm implements osFoundry and other AI for US medical device companies — focused on operational acceleration with validation and quality expectations in mind, and training your team. Pricing is fixed and public: a $399 assessment and $3,999/month implementation, with no per-seat fees. If you’d rather explore the platform first, go straight to osFoundry; if you want device-company AI done right, that’s where dgm comes in.