QAInsights has released AIx, an open standard and badge system that lets software developers self-declare the degree of AI involvement in their codebases. Using a 1–5 scale built around authorship metaphors — from "Verse" (fully human-written) to "Lorem Ipsum" (AI-generated output shipped largely unedited) — developers can embed a single badge in their project README to signal how much of the code reflects human intent versus AI generation. The standard is released under CC0 1.0, meaning it is fully public domain with no attribution required, and is explicitly framed as a transparency tool rather than a quality judgment.
AIx enters a crowded but inconsistent field of AI disclosure efforts in open source. Mitchell Hashimoto's Ghostty terminal emulator introduced an AI_POLICY.md requiring contributors to disclose the specific AI tools they used and demonstrate they understood the code they submitted. Researcher Melissa Weber Mendonça maintains a curated list of AI contribution policies from projects including Apache Airflow and IREE, and a RedMonk analysis from February 2026 mapped AI policies across more than 70 open source organizations. The critical distinction between AIx and these alternatives is scope: every existing framework—whether <a href="/news/2026-03-14-redox-os-adopts-no-llm-contribution-policy-amid-growing-oss-ai-governance-debate">strict bans on AI contributions</a> or disclosure mandates—governs contributor behavior through policy documents, while AIx applies a single spectrum score to the codebase itself, expressed as a badge signal any visitor can read at a glance.
The standard's honor-system nature is its defining principle — and yes, that also makes it easy to ignore. Institutional mandates have not yet reached source code authorship: the EU AI Act's Article 50 transparency obligations, effective August 2026, and California's SB-942 target AI-generated content outputs and consumer-facing systems, not software repositories. GitHub, npm, and PyPI have announced no metadata fields for AI code disclosure. This institutional vacuum means AIx faces no forcing function for adoption, but it also means the project occupies genuine uncontested territory. The voluntary standards that achieved lasting adoption — Semantic Versioning, Conventional Commits, the Contributor Covenant — did so through tooling integration and social visibility rather than mandates, a path AIx's lightweight implementation and zero-friction licensing are structured to follow.