AIx는 소프트웨어 프로젝트에 AI 참여를 공개하기 위한 개방형 표준입니다.
hackernews
|
|
📦 오픈소스
#ai 참여
#aix
#review
#개방형 표준
#소프트웨어
#저작권
원문 출처: hackernews · Genesis Park에서 요약 및 분석
요약
소프트웨어 프로젝트 내 AI의 참여 수준을 문학적 장르를 빗대어 5단계로 구분하여 공개하는 오픈 표준 'AIx'가 개발되었습니다. 이 표준은 코드를 문학 작품처럼 보고 AI의 활용 비중을 투명하게 점수화하며, 1점은 전적으로 인간이 작성한 경우, 5점은 생성된 그대로 배포된 경우를 의미합니다. 개발자는 README.md에 뱃지를 추가하거나 텍스트로 점수를 명시하여 AI 활용의 명확성과 투명성을 확보할 수 있습니다.
본문
Code is literature. How much of yours did you actually write? Quillx is an open standard for disclosing AI involvement in software projects - expressed through the language of authorship. Not a judgment. Just transparency. Badge Score Label Meaning 1 Verse Every line deliberate. Crafted like poetry. Human-authored entirely. 2 Prose Human voice throughout. AI helped with scaffolding and repetition. 3 Adapted A collaboration. Human-directed, AI-generated, meaningfully edited. 4 Ghostwritten AI wrote it. You briefed it, reviewed it, and signed off. 5 Lorem Ipsum Placeholder shipped as product. Generated, accepted, deployed. Add to your README.md : [ ![ Quillx ] ( https://raw.githubusercontent.com/qainsights/Quillx/main/badges/quillx-2.svg )] ( https://github.com/qainsights/Quillx ) Or declare it in plain text: Quillx: 2/5 · Prose - Architecture and logic are mine. AI scaffolded the boilerplate and tests. Code has style. It has voice. A codebase written with intent reads differently from one that was generated and accepted without scrutiny. The writing metaphor isn't decorative. It's accurate. Read the full SPEC.md for scoring criteria. Transparency over purity - Not a competition. Honest disclosure. Spectrum over binary - AI use is a gradient, not a switch. Self-declared - Trust the author. The community validates over time. Version your score - Projects evolve. Scores can too. CC0 1.0 Universal - free for any use, no attribution required. Started by QAInsights · Because code is literature.
Genesis Park 편집팀이 AI를 활용하여 작성한 분석입니다. 원문은 출처 링크를 통해 확인할 수 있습니다.
공유