|
Work item:
|
X.S-AIA
|
|
Subject/title:
|
Security requirements and guidelines for Artificial Intelligence agent
|
|
Status:
|
Under study
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2027-03 (Medium priority)
|
|
Liaison:
|
-
|
|
Supporting members:
|
-
|
|
Summary:
|
At present, artificial intelligence (AI) is undergoing a profound transformation, and as the core driving force of this transformation, the AI agent is gradually changing human’s life and work patterns. AI agent has the ability to perceive the environment, make autonomous decisions, and perform tasks, and its application areas are constantly expanding, affecting people's lifestyles and social operating mechanisms. However, the autonomy, interactivity and goal-orientation of AI agents, while bringing convenience, have also brought a number of application threats and challenges. The characteristic of AI agent brings special security threats and corresponding security requirements besides the general security requirements for AI.
However, the existing AI security standards are not yet able to provide a comprehensive and systematic guidance for the security application of AI agents. Therefore, in order to promote the healthy and sustainable development of AI agents, this recommendation focuses on the threat identification and proposes security requirements for the capabilities of AI agents, including perception & cognition, planning, memory and action, and then provides security measures as guidelines for the AI agent.
AI agent practitioners can use this recommendation to identify the threats and launch security protection for using, developing and operating AI agent.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2025-04-17 14:28:59
|
|
Last update:
2025-04-17 14:34:10
|
|
