Hello TSC members,

Artificial intelligence (AI) is increasingly used for code and documentation generation. In parallel, Europe is establishing an AI regulation, the EU AI Act , which requires transparency.

After discussion in the OpenAMP system reference call, we propose to use same approach already used in several projects, such as the Linux kernel and Zephyr:


In this context, we request a vote to approve the application of following AI policy:

When AI tools are used to contribute to the OpenAMP project, proper attribution helps track the evolving role of AI in the development process. Contributions should include an Assisted-by tag in the following format:

 

Assisted-by: AGENT_NAME:MODEL_VERSION [TOOL1] [TOOL2]

 

Where:

 

AGENT_NAME is the name of the AI tool or framework

MODEL_VERSION is the specific model version used

[TOOL1] [TOOL2] are optional specialized analysis tools used, for example, coccinelle, sparse, and smatch

Basic development tools (git, gcc, make, editors) should not be listed.

Example: Assisted-by: Claude:claude-3-opus coccinelle sparse



Let us know if you have any questions. 

 

Votes can be sent directly to me, with Nathalie and Bill in cc. The rules for the vote are the same applied for a Board voting, described in the charter.

Thanks and Regards
Arnaud

 

From: Arnaud POULIQUEN via Tsc <tsc@lists.openampproject.org>
Sent: Wednesday, June 17, 2026 9:02 AM
To: Chan King Choy, Nathalie <nathalie.chan-king-choy@amd.com>; OpenAMP TSC (tsc@lists.openampproject.org) <tsc@lists.openampproject.org>
Subject: [OA-Tsc] Re: Creating AI-generated code policy for OpenAMP

 

Hello Nathalie,

Thank you for the proposal. It seems to be a good approach. A chapter can be added to the README.md files, which already contain a contributing chapter.

Regards,

Arnaud

 

From: Chan King Choy, Nathalie via Tsc <tsc@lists.openampproject.org>
Sent: Wednesday, June 17, 2026 1:01 AM
To: OpenAMP TSC (tsc@lists.openampproject.org) <tsc@lists.openampproject.org>
Subject: [OA-Tsc] Creating AI-generated code policy for OpenAMP

 

AMD General

 

Hi OpenAMP TSC members,

 

The question came up in a recent OpenAMP System Reference call regarding use of AI generated code for OpenAMP.  Many companies are pushing heavily for their engineers to adopt AI assistance wherever possible.

 

Here are some resources to seed the discussion for creating AI-generated code policy for OpenAMP.

 

OpenAMP ties in closely to Linux kernel. Policy for Linux kernel: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/coding-assistants.rst

  • Human is responsible fully responsible for the contribution to the kernel, ensuring it follows the rules & meets license requirements
  • Transparency: Use assisted-by tag

 

OpenAMP ties in closely to Zephyr.  Policy for Zephyr is nearly identical to Linux kernel, like one did copy-paste-tweak of the other: https://docs.zephyrproject.org/latest/contribute/guidelines.html#ai-coding-assistants

 

In case folks want to broaden their analysis, here are some references shared by the Linux Foundation legal team to the LF Member Counsels:

    • Each Linux Foundation project gets to decide for themselves
    • High-level guidelines provided
    • Debian didn’t set an official policy & would handle case-by-case. 
    • Red Hat highlighted list of legal issues related to AI being debated in OSS communities

 

 

What are your thoughts on creating an AI-generated code policy for OpenAMP?

 

Thanks & regards,

Nathalie