Position: If Open Source Is to Win, It Must Go Public

Joshua Z Tan, Nicholas Vincent, Katherine Elkins, Magnus Sahlgren, Joseph Low, David Huu Pham, Sampo Pyysalo, Jenia Jitsev
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:172248-172261, 2026.

Abstract

Open source projects have made incredible progress in producing widely usable machine learning models and systems, but open source alone will face challenges in fully democratizing access to AI. Unlike previous generations of open source software, open source and open weight AI models require substantial resources to activate and maintain—e.g., data and compute for pre-training, post-training, and deployment—which only a few actors can currently provide. This position paper argues that open source AI must be complemented by public AI: infrastructure and institutions that ensure models are accessible, sustainable, and governed in the public interest. To achieve the full promise of AI models as prosocial public goods, we need to build public infrastructure to power and deliver open source software and models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-tan26i, title = {Position: If Open Source Is to Win, It Must Go Public}, author = {Tan, Joshua Z and Vincent, Nicholas and Elkins, Katherine and Sahlgren, Magnus and Low, Joseph and Pham, David Huu and Pyysalo, Sampo and Jitsev, Jenia}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {172248--172261}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/tan26i/tan26i.pdf}, url = {https://proceedings.mlr.press/v306/tan26i.html}, abstract = {Open source projects have made incredible progress in producing widely usable machine learning models and systems, but open source alone will face challenges in fully democratizing access to AI. Unlike previous generations of open source software, open source and open weight AI models require substantial resources to activate and maintain—e.g., data and compute for pre-training, post-training, and deployment—which only a few actors can currently provide. This position paper argues that open source AI must be complemented by public AI: infrastructure and institutions that ensure models are accessible, sustainable, and governed in the public interest. To achieve the full promise of AI models as prosocial public goods, we need to build public infrastructure to power and deliver open source software and models.} }
Endnote
%0 Conference Paper %T Position: If Open Source Is to Win, It Must Go Public %A Joshua Z Tan %A Nicholas Vincent %A Katherine Elkins %A Magnus Sahlgren %A Joseph Low %A David Huu Pham %A Sampo Pyysalo %A Jenia Jitsev %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-tan26i %I PMLR %P 172248--172261 %U https://proceedings.mlr.press/v306/tan26i.html %V 306 %X Open source projects have made incredible progress in producing widely usable machine learning models and systems, but open source alone will face challenges in fully democratizing access to AI. Unlike previous generations of open source software, open source and open weight AI models require substantial resources to activate and maintain—e.g., data and compute for pre-training, post-training, and deployment—which only a few actors can currently provide. This position paper argues that open source AI must be complemented by public AI: infrastructure and institutions that ensure models are accessible, sustainable, and governed in the public interest. To achieve the full promise of AI models as prosocial public goods, we need to build public infrastructure to power and deliver open source software and models.
APA
Tan, J.Z., Vincent, N., Elkins, K., Sahlgren, M., Low, J., Pham, D.H., Pyysalo, S. & Jitsev, J.. (2026). Position: If Open Source Is to Win, It Must Go Public. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:172248-172261 Available from https://proceedings.mlr.press/v306/tan26i.html.

Related Material