The following is a summary of the governing structures and policies for the program and projects of the AI Alliance
Outline
Definitions
A Project in the AI Alliance consists of an organized set of work with clear goals, governance, and openly accessible output artifacts. This can include anything from a data set to a software stack to a position paper about an aspect of AI policy.
All projects are subject to straightforward but important requirements on openness, transparency, and governance. These are detailed in the “Projects and IP” section of this document.
Supported Projects (formerly some “affiliated projects”) are those that remain governed by a Member of the AI Alliance, but for which the AI Alliance can choose to provide specific elements of support. Examples of such governing organizations include individual companies, non-profits or open source foundations. Examples of this support include recipes and developer guidance, community management, red teaming, marketing and awareness, etc.
Supported Projects are subject to minimum requirements on transparency, contributor opportunity and IP, permissive use licensing, and community conduct.
Compliance with these requirements is managed by the governing organization(s) with review by the AI Alliance, with the AI Alliance having the right to end or change its support at its discretion. Typically, a Supported Project will be listed in the AI Alliance Github organization and website to facilitate collaboration and community engagement, but the project itself including artifacts, issues and project management, will be controlled and run by the AI Alliance Member.
Managed Projects (formerly “core projects”) are those that are governed and run by the AI Alliance legal entity, with defined project leadership within the AI Alliance program.
Managed Projects are subject to the same minimum requirements on transparency, contributor opportunity and IP, permissive use licensing, and community conduct as Supported Projects.
Compliance with these requirements is managed by the AI Alliance, typically through delegation to a specific project leadership team.
A Managed Project will live in the AI Alliance Github organization, including artifacts, issues and project management, and community engagement.
A Managed Project will come under the oversight of either the 501(c)3 or the 501(c)6 depending on the nature of the project, its goals and funding model.
Member Projects (formerly most “affiliated projects”) are projects that are aligned with the AI Alliance program goals of the AI Alliance and remain fully managed and supported by individual AI Alliance members. All Affiliated Projects must have open permissively licensed artifacts and clear community calls-to-action.
Individuals who are responsible for goals, maintainership, decisions, resources, or other aspects of specific projects on a day-to-day basis.
Any individual person (for clarity, including the actions of an AI agent or system owned by or acting under the oversight of an individual person) that contributes to any project in nearly any substantive way. Collaborators can be from member organizations or the broader community. Collaborators are not just users.
An organization that has committed to be an anchor sponsor of a major component of the AI Alliance program and is part of the AI Alliance overall governance based on the defined membership process.
A group of collaborators working together toward a specific goal for a project or initiative. Typically a working group will be part of a Focus Area, but it may cut across more than one focus areas.
We take a streamlined approach and seek not to over-define structures and terms. As needed to advance the program, other terms may be used to describe elements of projects and programs, including Focus Area, Initiative, Sponsor, etc. A Focus Area or Initiative is a grouping of projects under a common theme. A Sponsor is any organization or person that makes a significant resource commitment to a project or initiative. This may include people, funding, computers, data, or other services.
Boards and Committees
The AI Alliance aims to accelerate and disseminate open innovation across the AI technology landscape to improve foundational capabilities, safety, security and trust in AI, and to responsibly maximize benefits to people and society everywhere. This is a broad heterogeneous program that embraces diversity of thought and action. As such we favor a lightweight operating and governing structure with maximum empowerment of individual collaborators and organizational members and sponsors to pursue projects in ways best fit to the project. The processes and governance of AI Alliance are intended to support this.
The AI Alliance is made up of two closely linked non-profit legal entities. Each of these entities has a governing board to decide matters of overall program operations including finance and legal matters.
- A US 501(c)3 charitable non-profit entity focused on coordinating research and experimentation, and disseminating AI software, data, models and tools that anyone can contribute to, test, and experiment with to improve the safety, trust and accuracy of AI, and educate on the use of AI.
- A US 501(c)6 trade association focused on specific industry-focused work such as policy lobbying, and establishing commercial practices or standards.
Each of these entities has a governing board (“Board”) to decide matters of overall program operations including finance and legal matters. The primary role of the AI Alliance Boards is to run, including structuring and delegating, the business operations of the AI Alliance and provide overall program oversight. This includes centralized legal, compliance, IP, finance, contracting, procurement, tooling, process definition and the delegation of these responsibilities to the relevant committee, including but not limited to the operating committee.
The AI Alliance boards are generally not involved in day to day project-level or working group-level decisions.
Project Governance
Projects are the heart of the AI Alliance and how we achieve our goals and impact.
We aim to empower founders and project leaders with a lightweight onboarding and governance process that ensures they maintain technical leadership of the project while providing clear support and a framework to help them scale that is purpose-built for open source AI and for the needs of the AI community. Our requirements seek to ensure openness and derisk the project’s adoption by users and organizations.
Projects come in three types: Supported Projects, Managed Projects, and Member Projects.
The process enablement of project onboarding and lifecycle management is a responsibility of the Operations Committee. The requirements are determined by the Boards.
The addition or exit of Managed Projects and Supported Projects must be reviewed and approved by the relevant Board ((c)3 or (c)6) or by delegate(s) of the Boards. The delegate(s) can be an ad hoc committee established by the Boards, the Steering Committee or specific working groups associated with initiatives or focus areas.
The addition or exit of Member Projects are determined by the Member and the Operating Committee as part of its Membership management responsibility.
The AI Alliance requires all projects to have:
- Transparency on who are the Project Leaders.
- Transparency on how decisions and contributions to the project are made, and especially for Managed Projects that these decisions are community-driven.
- Transparency on the origin of code, data, and funding contributions.
- Clarity on how a person can contribute.
- A standard contributor Developer Certificate of Origin (DCO) or similar, providing IP rights to the project while preserving IP ownership of the contributing individual.
- A permissive use license for the project artifacts and open availability of the artifacts.
- A broader set of licenses for Supported Projects that are foundation models is allowed, because of the nascent and evolving state of model licenses and because of the widespread popularity of safe and responsible use terms.
- Specific licenses / terms that are accepted are decided at the project intake process by the operations committee of the AI Alliance working with the relevant working group leads.
- For Managed Projects that are models the AI Alliance uses the Apache 2.0 license at this time.
- For non-model artifacts, standard OSDI-compatible licenses are required for all.
- A commitment to open, safe, secure and responsible community development via the AI Alliance Community Code of Conduct and a commitment to the AI Alliance Non-competition rules is required of all Project Leaders.
Supported Projects must be governed by an AI Alliance member organization. This member makes an organizational-level public commitment to the requirements as part of their membership and project onboarding. If the requirements are breached for whatever reason, the AI Alliance reserves the right to rally the community and its resources to provide a remedy. This remedy could take various forms – we desire the simplest, lowest friction remedy possible. Remedies will be reviewed and approved by the Board(s) or its delegate(s).
When projects are submitted for consideration to the AI Alliance as Supported Projects, all projects will have minimum requirements to satisfy. Beyond this, the project organizers can request/opt-in to additional support from the AI Alliance on an as needed basis if both the AI Alliance (through the relevant working group and/or steering committee) and the project leadership agree. AI Alliance Managed Projects represent a “full” opt-in, where the AI Alliance is responsible for all aspects of the project support and governance.
Community Management
A vibrant, open global community is how new ideas arise and projects are built, grow and achieve impact. Management of the growth and engagement of the AI Alliance community will be for the purpose of achieving project goals. Driving direct project-level impact metrics including but not limited to awareness, downloads and use, contribution, and production adoption are the primary focus.
Community management is in general associated with projects, and/or Focus Areas or Initiatives of projects. Community activity including events, content and communication will be in service to projects.
Community management will be overseen by Project Leaders and the board(s) or delegates of the boards.
All individuals participating in the AI Alliance Community must adhere to the AI Alliance Code of Conduct.
Intellectual Property
Contribution of IP to an AI Alliance Managed Project does not require or involve the contributor to give away the contributor’s ownership in the contributed IP. The Developer Certificate of Originality (DCO) used by the AI Alliance does not change ownership of the contribution (e.g. the AI Alliance does not become the “owner” of contributions”), but instead requires that anyone submitting content to a Managed Project affirm that they can and are providing anyone and everyone a license to their contribution under the applicable open source license(s) for the project. This is true regardless of whether the contribution is submitted by an individual from a member or non-member of the AI Alliance.
Contribution of IP to an AI Alliance Supported Project in general follows different processes, which are typically outlined in the relevant Github repository, web or other documentation associated with the project, subject to the requirements above.
The AI Alliance has standard open source licenses that it applies depending on the type of artifact contributed to the Managed Project, and a Managed Project should not modify these standard licenses without exception approval by the Board or its delegate(s). These licensees are:
- Code: Apache 2.0; “code” refers to code, software, or other similar content written in a computer language.
- Data: CDLA Permissive 2.0; information or other content contributed for the purpose of being part of a dataset.
- AI Model Weights: Apache 2.0; parameters, weights, and biases in an AI model.
- Documentation and similar content: CC BY 4.0; information or other content that is not Code, Data, or AI Model Weights contributed for the purpose of informing persons who read it.
For example, an AI project model would include model weights and associated software under Apache 2.0, with any model card or other information about the model and its performance under the CC BY 4.0.
We acknowledge that certain artifacts may call for a deviation from these standard licenses, for example if a particular concern about safe use is identified. These situations are handled by the exception process through the Board(s) or its delegate(s).
Projects involving data should keep in mind that the provenance, dependencies and license terms of the source(s) of data, and relevant legal and regulatory statutes may affect the ability to contribute. For example, unlike open source software, it is rare for data publicly available on the Internet to be made available for use under a license. Whenever there is a question about ownership and usage of data, please seek guidance from the relevant lead(s)/maintainer(s) of the project before proceeding. In particular, the AI Alliances Open Trusted Data Initiative was established to address the many challenges and opportunities in data for AI.
Confidential information is at odds with open technology and open innovation and should not be exchanged or contributed as part of AI Alliance Core Projects, and more broadly should never be introduced into any Alliance project, discussion or event. Avoiding exchanging confidential information also supports antitrust compliance as laid out by the Competition Law Guidelines for the AI Alliance.
While we aspire to have all work and all output of AI Alliance be open, accessible and permissively usable at all times, we appreciate that work in progress may not always be amenable to incremental release. In these scenarios we still encourage collaborators to release early and often, use Github and publicly-accessible documents to ensure broad access, and use reasonable effort to invite the broad community to participate when feasible.
Managed Project repositories should include links to the AI Alliance’s Code of Conduct and the Competition Law Guidelines for the AI Alliance. Leads should reference Competition Law Guidelines for the AI Alliance in formal meetings by reading or showing the following language:
An important reminder is that all AI Alliance activities need to comply with applicable antitrust and competition laws and that you should comply with the Competition Law Guidelines for the AI Alliance, which are available from the AI Alliance website. There should be no discussions between competitors enabling agreements or concerted actions that restrain competition, including the exchange of confidential information that enables competitors to engage in illegal concerted action. Moreover, participation in AI Alliance activities does not commit any participant to use any specific technology, restrict the use or sale of any technology, or otherwise limit any participant’s ability to engage in similar activities with others. If you have questions about these matters, please contact your legal counsel for advice.
If you have questions about this guidance, please reach out to the relevant lead.