arbiter
JournalsAccount

The case for contextual copyleft: licensing open-source training data and generative AI

Grant Shanklin et al.

International Journal of Law and Information Technology2026https://doi.org/10.1093/ijlit/eaag003article
ABDC A
Weight
0.50

Abstract

The rise of generative AI systems presents new challenges for the Free and Open-Source Software (FOSS) community, particularly around applying copyleft principles when open-source code is used to train AI models. This article introduces the Contextual Copyleft AI (CCAI) licence, a novel use of the copyleft mechanism that extends licence obligations from training data to resulting generative models. The CCAI licence enhances developer control, incentivizes open-source AI, and mitigates open-washing. A structured three-part evaluation examines: (i) legal feasibility under current copyright law, (ii) policy justification across traditional software and AI, and (iii) cross-contextual benefits and risks. Still, open-source AI carries a higher risk—especially misuse—making complementary regulation essential to achieve a fair risk-benefit balance. The article concludes that, within a robust regulatory environment focused on responsible AI, the CCAI licence offers a viable path for preserving and adapting core FOSS values to meet the demands of modern AI development.

Open via your library →

Cite this paper

https://doi.org/https://doi.org/10.1093/ijlit/eaag003

Or copy a formatted citation

@article{grant2026,
  title        = {{The case for contextual copyleft: licensing open-source training data and generative AI}},
  author       = {Grant Shanklin et al.},
  journal      = {International Journal of Law and Information Technology},
  year         = {2026},
  doi          = {https://doi.org/https://doi.org/10.1093/ijlit/eaag003},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

The case for contextual copyleft: licensing open-source training data and generative AI

Flags are reviewed by the Arbiter methodology team within 5 business days.

Evidence weight

0.50

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.50 × 0.4 = 0.20
M · momentum0.50 × 0.15 = 0.07
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.