Skip to content

Create ai_policy.md#24

Closed
jmccarten-wetafx wants to merge 1 commit intoai_policyfrom
jmccarten-wetafx-patch-1
Closed

Create ai_policy.md#24
jmccarten-wetafx wants to merge 1 commit intoai_policyfrom
jmccarten-wetafx-patch-1

Conversation

@jmccarten-wetafx
Copy link
Copy Markdown

No description provided.

Signed-off-by: John McCarten <jmccarten@wetafx.co.nz>
Comment thread tsc/process/ai_policy.md

A human is required.

PRs generated without an identifiable human author will be immediately closed and the source account banned from the project.
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Banned is possibly too strong, soften the terms, give a warning potentially, hold the PR back for further review. "We reserve the right to ban accounts if repeated or egregious violation on a case by case basis".

Apply to any violation not just AI policy.

Reword to assume good intent?

Comment thread tsc/process/ai_policy.md

If AI was used, it is *mandatory* that the commit must disclose via 'Assisted-by' followed by the tool used and a brief description of what it was used for:

```Assisted-by: anthropic-claude-connet-4.6 - Generated tests for Model input processing```
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sonnet not connect

Comment thread tsc/process/ai_policy.md

## Disclosure

If AI was used, it is *mandatory* that the commit must disclose via 'Assisted-by' followed by the tool used and a brief description of what it was used for:
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Commit hooks to check for format - possibly validate model names.
Consider any use of AI even if not generative?
Mandatory to disclose if artifact makes it into repo, optional disclosure if use of AI for introspection or non generative tasks.

Comment thread tsc/process/ai_policy.md

Regardless of how the code was created - only humans can submit PRs. Only humans can review PRs. Obly humans can commit them to the repository.

A human is required.
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A human should understand what is in the code and how it works (avoid pure vibe coding).
There should be a trail of involvement of a human in the development - specifications, plans, slack discussions.

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Updates to issues as an example of involment

Comment thread tsc/process/ai_policy.md

## Graduated adoption

RMTC is an early stage project - it's code and community examples not yet established enough to allow for an AI to generate complementary code. Additionally, it is not clear regarding the possibility for reconstruction of license violating material for burgenoning projects which poses a specific issue for RMTC due to it's aims for responsible AI usage.
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
RMTC is an early stage project - it's code and community examples not yet established enough to allow for an AI to generate complementary code. Additionally, it is not clear regarding the possibility for reconstruction of license violating material for burgenoning projects which poses a specific issue for RMTC due to it's aims for responsible AI usage.
RMTC is an early stage project - its code and community examples not yet established enough to allow for an AI to generate complementary code. Additionally, it is not clear regarding the possibility for reconstruction of license violating material for burgenoning projects which poses a specific issue for RMTC due to its aims for responsible AI usage.

Comment thread tsc/process/ai_policy.md

## Graduated adoption

RMTC is an early stage project - it's code and community examples not yet established enough to allow for an AI to generate complementary code. Additionally, it is not clear regarding the possibility for reconstruction of license violating material for burgenoning projects which poses a specific issue for RMTC due to it's aims for responsible AI usage.
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Phase 2-5 a little fine grained, maybe make more general - what is 'minor' vs 'major'?

Consider difference between policy and application - case by case?
Important for gradual adoption for greenfield project.
A little control is good.
These are intended to guide our AI usage not hard rules

Consider in Phase 1

Comment thread tsc/process/ai_policy.md

## AI usage definition

AI has been used for sometime in software development, from code suggestions to spell checking. What does this policy cover?
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

'some time' is a period of time; 'sometime' is a specific point in time

Suggested change
AI has been used for sometime in software development, from code suggestions to spell checking. What does this policy cover?
AI has been used for some time in software development, from code suggestions to spell checking. What does this policy cover?

Comment thread tsc/process/ai_policy.md

- Using it to query, summarise or review the code but not generate an artifact for PR
- Spell checking
- Traditional line code completion (<10lines)
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Remove lines reference - remove line counts as too arbitrary

@vvalderrv vvalderrv closed this Apr 9, 2026
@vvalderrv vvalderrv deleted the jmccarten-wetafx-patch-1 branch April 9, 2026 17:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants