IPAe embeds the IoT Profile Assistant within the eSIM itself, while IPAd places that functionality in the device operating system. The distinction matters because IPAe can reduce firmware change requirements and simplify lifecycle management, whereas IPAd can create more integration, testing, and certification work for OEMs and suppliers.
What changes when IPAe keeps the IoT Profile Assistant in the eSIM?
IPAe makes the eSIM the place where profile-assistant logic lives, so provisioning and lifecycle actions can stay closer to the subscription and profile material they govern. That usually reduces how much the device firmware or operating system has to change, which is useful when many device variants must be supported or when certification cycles are expensive.
That design also shifts some integration responsibility toward the eSIM and its surrounding provisioning ecosystem, because the assistant behavior is no longer primarily delivered by the device platform. The practical trade-off is simpler device-side change management, but tighter dependency on eSIM capability, vendor implementation quality, and interoperability across the activation flow.
The architecture choice is less about abstract preference and more about where you want the lifecycle control point to sit. If the assistant must travel with the eSIM and remain consistent across device swaps, IPAe is the more natural fit. If the device environment is expected to host the logic locally and integrate it with other OS services, the operational burden moves to the handset or endpoint platform instead.
How IPAd changes the device and supplier burden
IPAd places the IoT Profile Assistant in the device operating system, so the endpoint becomes the execution environment for profile handling. That can fit platforms that already expose the right APIs, trust anchors, and update cadence, but it also means the OEM, OS integrator, and certification path matter much more than they do in an eSIM-hosted model.
Because the device owns more of the workflow, IPAd tends to increase testing scope. Teams have to validate the OS integration, device-specific behavior, update compatibility, and edge cases around activation, profile switching, and recovery. In practice, this is where fragmentation appears, especially when a product line spans multiple chipsets, firmware baselines, or regional variants.
IPAd can still be the right answer when the device platform needs explicit control over user experience or system integration. The key point is that the architectural convenience of local hosting comes with a wider operational surface for OEMs and suppliers, particularly during certification and long-term maintenance.
Why the distinction matters in eSIM IoT architectures
The difference between IPAe and IPAd is a placement decision, but the downstream effects are architectural. IPAe centralises assistant functionality in the eSIM and can make the endpoint thinner, while IPAd distributes that functionality into the device stack and makes the endpoint more responsible for profile operations. That affects portability, upgrade strategy, device validation, and who owns failures when activation does not behave as expected.
It also changes how teams think about resilience and lifecycle management. An eSIM-centric model can simplify device refresh and reduce firmware churn, while a device-centric model may align better with tightly integrated products that already have mature OS support. For readers working through standards or governance questions, the relevant source of truth is the GSMA eSIM architecture documentation, which defines how the IoT Profile Assistant is placed and used across these models.
At a practical level, the choice usually comes down to whether you want the profile-assistant function to be portable with the subscription or embedded with the endpoint. That decision has consequences for supplier coordination, supportability, and how easily a deployment can absorb changes over time.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
NIST SP 800-53 Rev 5, NIST CSF 2.0 and CSA Cloud Controls Matrix set the technical controls, while ISO/IEC 27001:2022 defines the regulatory obligations.
| Framework | Control / Reference | Relevance |
|---|---|---|
| NIST SP 800-53 Rev 5 | IA-9 — Service Identification and Authentication | eSIM IoT assistants rely on device-to-service trust and authenticated interactions. |
| Recommendation — Use IA-9 to authenticate device-side and service-side profile interactions. | ||
| ISO/IEC 27001:2022 | A.5.15 — Access control | IPAe versus IPAd changes where profile access and control are enforced. |
| Recommendation — Define and enforce which component may initiate profile-assistant actions. | ||
| NIST CSF 2.0 | PR.AA-05 — Identity Management, Authentication and Access Control | The architecture shifts access control responsibility between eSIM and device OS. |
| Recommendation — Assign clear control ownership for profile-assistant access decisions. | ||
| CSA Cloud Controls Matrix | IAM — Identity and Access Management | eSIM IoT architectures depend on who governs identity and access in the activation flow. |
| Recommendation — Map assistant placement to the IAM boundary that owns lifecycle control. | ||
Practitioner Guidance
What to verify: Confirm whether the device fleet can support the certification, update, and recovery burden of a device-hosted assistant before choosing IPAd. If the programme spans multiple OEMs or long device lifetimes, test the operational cost of fragmentation early rather than treating it as an integration detail.
Decision rule: If the main objective is to minimise device-side change and preserve portability across hardware swaps, prefer an eSIM-hosted model. If the main objective is deep OS integration and the platform already owns the relevant lifecycle hooks, IPAd may be acceptable, but only with clear ownership for testing and ongoing maintenance.
Practitioner takeaway: The real decision is not where the feature sounds cleaner, but where you want the lifecycle control point and who can sustain the integration burden over the full device life.
Related resources from NHI Mgmt Group
- What is the difference between privilege reduction and secret rotation?
- What is the difference between a rules-based secret scanner and a hybrid scanner?
- What is the difference between code scanning and runtime identity monitoring?
- What is the difference between zero trust for users and zero trust for NHIs?
Deepen Your Knowledge
Reviewed and updated by the NHIMG editorial team on September 29, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org
