|
Work item:
|
X.sf-VIoT
|
|
Subject/title:
|
Security framework for vertical IoT devices using multimodal AI systems
|
|
Status:
|
Under study
|
|
Approval process:
|
TAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2028-10 (Medium priority)
|
|
Liaison:
|
ITU-T SG20, ITU-T SG21, ISO/IEC JTC1 SC27
|
|
Supporting members:
|
Korea Rep. of), KISA, China mobile, ETRI
|
|
Summary:
|
Vertical Internet of Things (IoT) devices are rapidly evolving from primarily focused on simple sensing and communication functions into autonomous cyber-physical systems equipped with multimodal artificial intelligence (AI). In vertical sectors such as smart home, smart manufacturing, autonomous healthcare, automotive, and smart buildings, these devices increasingly combine heterogeneous sensors (e.g., vision, audio, radar, LiDAR, telemetry) with AI models. However, this kind of evolution also introduces a substantially broader attack surface.
Therefore, this draft Recommendation provides a comprehensive security framework for vertical Internet of Things (IoT) devices using multimodal artificial intelligence (AI). It defines a closed-loop operational architecture comprising perception, world model, planning, control, action, and feedback loop layers. It also analyzes security threats and vulnerabilities across each layers and define corresponding security capabilities and security requirements. The draft Recommendation is applicable to vertical IoT deployments in domains such as smart home, smart manufacturing, autonomous healthcare, automotive, smart buildings, and other safety-relevant cyber-physical environments.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2026-06-10 12:10:11
|
|
Last update:
2026-06-10 12:31:02
|
|
