Today’s most advanced AI systems and foundation models are largely proprietary and controlled by a small number of companies. There is a striking lack of viable public or open alternatives. This gap means that cutting-edge AI remains in the hands of a select few, with limited orientation toward the public interest, accountability or oversight.
Public AI is a vision of AI systems that are meaningful alternatives to the status quo. In order to serve the public interest, they are developed under transparent governance, with public accountability, equitable access to core components (such as data and models), and a clear focus on public-purpose functions.
This white paper—commissioned by Bertelsmann Stiftung through their reframe[Tech] – Algorithms for the Common Good project and authored by Felix Sieker, Alek Tarkowski, Lea Gimpel, and Cailean Osborne—builds on earlier proposals for Public AI and is aimed at policymakers and funders, with the goal of helping to turn the vision of Public AI into reality. In particular, it advances this timely conversation by making the following two novel contributions.
A vision for Public AI grounded in the reality of the AI stack
A vision for public AI needs to take into account today’s constraints at the compute, data and model layers of the AI stack, and offer actionable steps to overcome these limitations. This white paper offers a clear overview of AI systems and infrastructures conceptualized as a stack of interdependent elements, with compute, data and models as its core layers.
It also identifies critical bottlenecks and dependencies in today’s AI ecosystem, where dependency on dominant or even monopolistic commercial solutions constrains development of public alternatives. It highlights the need for policy approaches that can orchestrate resources and various actors across layers, rather than attempting complete vertical integration of a publicly owned solution.
To achieve this, it proposes three core policy recommendations:
- Develop and/or strengthen fully open source models and the broader open source ecosystem
- Provide public compute infrastructure to support the development and use of open models
- Scale investments in AI capabilities to ensure that sufficient talent is developing and adopting these models
In order to achieve this, complementary pathways for Public AI development need to be pursued, focused on the three core layers of the AI stack: compute, data, and models:
- Compute Pathway: It focuses on providing strategic public computing resources, particularly supporting open-source AI development. Key recommendations include ensuring computing access for fully open projects, expanding compute for research institutions, and improving coordination between public compute initiatives.
- Data Pathway: It emphasizes creating high-quality datasets as digital public goods through commons-based governance. This includes developing datasets as publicly accessible resources while protecting against value extraction, and establishing public data commons with appropriate governance mechanisms.
- Model Pathway: It centers on fostering an ecosystem of fully open source AI models, including both a state-of-the-art “capstone model” and specialized smaller models. The strategy emphasizes building sustainable open source AI development capabilities rather than simply competing with commercial labs.
The “gradient of publicness”: A framework for Public AI
The white paper also offers a “gradient of publicness” framework, rooted in Public Digital Infrastructure principles. This framework can guide decision-making around investments in AI infrastructure and help increase public value while acknowledging existing constraints and limitations to building fully Public AI.
This framework maps AI interventions along a continuum – from fully public to fully private – based on their attributes (e.g. accessibility, openness, interoperability), functions (e.g. enabling social or economic goals) and modes of control (e.g. democratic governance and accountability). It serves as both a diagnostic and strategic tool for assessing where an intervention falls along this continuum, and for identifying interventions that could strengthen its public value.
