ANNUAL REPORT 2024
Partnership on AI (PAI) is a nonprofit community of academic, civil society, and industry organizations who come together to address the most important questions for the future of AI. In 2024, PAI and our Partners focused on real-world solutions for the responsible, safe, and inclusive development, deployment, and use of AI systems.
Aligned with its mission to ensure AI advances benefit people and society, PAI investigated AI’s impact on digital media and information, inspired evidence-based AI policy, leveraged AI for positive social impact, advanced safety in AI, and contributed to creating an inclusive economic future in AI. Our work in 2024 brought us several steps closer to realizing our vision of a future where AI is developed to empower humanity by contributing to a more just, equitable, and prosperous world.
From Our CEO
Since our founding, PAI has worked with our community of academic, civil society, and industry partners to jointly develop real-world guidance for the safe and responsible development, deployment, and use of AI. Together, we developed a clear, collective agenda for action in response to questions like: When and how should AI-generated media be labeled? How can workers share in the benefits of AI? How can frontier and foundation models be developed and deployed safely? How can citizens and communities benefit from AI?
In 2024, we worked with our Partners to take action on this agenda, driving impact in practice, policy, and public understanding.
It is with great pride in our community that I look back on the year and see our mission in action. We fostered changes in industry practice, informed emerging public policy, and advanced public understanding. Through case studies, we saw how partners like Adobe, OpenAI, and the BBC are implementing our Synthetic Media Framework. Government agencies from NIST in the US to the OECD have noted PAI’s guidance as inputs to public policies.
In addition, we continue to launch new initiatives in areas that most need the attention of our multistakeholder community. This year, we launched a new Philanthropy Steering Committee to better understand the opportunities and risks of AI to this sector. We mapped the value chain for open foundation models identifying key points of intervention to improve safety. We also released our first policy report with recommendations to improve policy interoperability. PAI’s work continues to address the biggest questions about the future of this transformative technology.
As always, our work is grounded in our global community of partners. Bringing together stakeholders from across sectors at in-person forums, workshops, and through online gatherings, we connected with more than 1000 experts within our community and grew to 128 PAI Partners across 16 countries.
I am grateful to work with this community, along with our supportive Board, committed donors, and talented staff. Thank you to all who have contributed and helped drive impact in 2024. I look forward to continuing this momentum in 2025 and beyond.
Rebecca Finlay
CEO
Investigating AI’s Impact on Digital Media and Information
In 2024, a year where AI-manipulated media was expected to play a role in elections taking place across the globe, PAI continued its work to promote truth and transparency across the digital media and information ecosystem.
Early in the year, PAI launched a cross-sector Community of Practice to explore different approaches to the challenges posed by the use of AI tools in elections, creating space for shared learning.
Demonstrating the application of PAI’s Synthetic Media Framework, PAI published 16 in-depth case studies from AI-developing companies like Adobe and OpenAI, media organizations such as CBC and BBC, platforms such as Meta and Microsoft, as well as civil society organizations like Thorn and WITNESS. The case studies were a requirement of Framework supporters, who explored how the best practices on responsible development, creation, and sharing of AI generated media could be applied to real-world use cases.
The first set of case studies from framework supporters and accompanying analysis, focused on transparency, consent, and harmful/responsible use cases. The second set of case studies from framework supporters focused on an underexplored area of synthetic media governance: direct disclosure — methods to convey to audiences how content has been modified or created with AI, like labels or other signals — PAI developed policy recommendations based on insights from the cases. If responsible synthetic media best practices, such as disclosure, are not implemented alongside safety recommendations for open source model builders, synthetic media may lead to real-world harm, such as manipulating democratic and political processes.
As AI-generated and manipulated media becomes commonplace, PAI acted as a catalyst for its community to explore and adopt responsible practices.
In 2024, a year where AI-manipulated media was expected to play a role in elections taking place across the globe, PAI continued its work to promote truth and transparency across the digital media and information ecosystem.
Early in the year, PAI launched a cross-sector Community of Practice to explore different approaches to the challenges posed by the use of AI tools in elections, creating space for shared learning.
Demonstrating the application of PAI’s Synthetic Media Framework, PAI published 16 in-depth case studies from AI-developing companies like Adobe and OpenAI, media organizations such as CBC and BBC, platforms such as Meta and Microsoft, as well as civil society organizations like Thorn and WITNESS. The case studies were a requirement of Framework supporters, who explored how the best practices on responsible development, creation, and sharing of AI generated media could be applied to real-world use cases.
The first set of case studies from framework supporters and accompanying analysis, focused on transparency, consent, and harmful/responsible use cases. The second set of case studies from framework supporters focused on an underexplored area of synthetic media governance: direct disclosure — methods to convey to audiences how content has been modified or created with AI, like labels or other signals — PAI developed policy recommendations based on insights from the cases. If responsible synthetic media best practices, such as disclosure, are not implemented alongside safety recommendations for open source model builders, synthetic media may lead to real-world harm, such as manipulating democratic and political processes.
Technical Program Manager, Policy Research
OpenAI
Inspiring Evidence-Based AI Policy
By convening experts across sectors, building consensus-based frameworks, and facilitating global policy dialogue, PAI has been able to inspire and support the development and implementation of impactful AI policy. Throughout 2024, PAI’s work was cited in various policy guidance and initiatives, demonstrating the impact of our resources, research and recommendations to inspire innovations in public policy.
Coordination is critical to developing and inspiring good AI policy and without it, we risk creating an inconsistent patchwork of frameworks and divergent understandings of best practices. Ensuring these frameworks work together is critical which is why we published the Alignment on AI Transparency — a comparative analysis of eight leading policy frameworks for foundation models, with a particular focus on documentation requirements, critical components of transparency and safety.
To further foster collaboration towards responsible AI policy, PAI hosted its AI Policy Forum in New York ahead of UN week in September. Bringing together representatives from the United Nations and national governments, alongside PAI’s community of academic, civil society, and industry partners, the Forum drove in-depth conversations about the need for people-centric policies.
Coordination is critical to developing and inspiring good AI policy and without it, we risk creating an inconsistent patchwork of frameworks and divergent understandings of best practices. Ensuring these frameworks work together is critical which is why we published the Alignment on AI Transparency — a comparative analysis of eight leading policy frameworks for foundation models, with a particular focus on documentation requirements, critical components of transparency and safety.
To further foster collaboration towards responsible AI policy, PAI hosted its AI Policy Forum in New York ahead of UN week in September. Bringing together representatives from the United Nations and national governments, alongside PAI’s community of academic, civil society, and industry partners, the Forum drove in-depth conversations about the need for people-centric policies.
Assistant Secretary of Commerce for Communications and Information
and NTIA Administrator
Leveraging AI for Positive Social Impact
Launch of PAI’s Philanthropy Steering Committee
PAI recognizes that philanthropic organizations play a critical role in driving change towards equitable AI and by leveraging data-driven approaches, these organizations can support the development and use of safe, responsible, and inclusive AI tools and systems. That is why in 2024 PAI launched the AI & Philanthropy Steering Committee, a pivotal new initiative fostering collaboration, dialogue, and action at the intersection of AI and philanthropy. The primary purpose of the AI & Philanthropy Steering Committee is to define and guide philanthropic initiatives that leverage AI for positive social impact. The committee dedicates its time to the broad spectrum of challenges and opportunities AI presents to the philanthropic and nonprofit landscape, comprising a diverse assembly of industry, philanthropy, nonprofit community leaders, academia and subject matter experts.
Philanthropy Steering Committee Members
Chan Zuckerberg Initiative
New America
Omidyar Network
Good Tech Advisory
Surdna Foundation
Google.org
Markkula Center for Applied Ethics
Data & Society
MetLife Foundation
Skoll Foundation
Surdna Foundation
AI Now Institute
Ford Foundation
Cohere For AI
Annie E. Casey Foundation
Ada Lovelace Institute
AI Collaborative
Amazon Web Services
Philanthropy Forum
PAI brought together the AI & Philanthropy Steering Committee and broader community of PAI Partners at theAI & Philanthropy Forum in March 2024. The forum fostered meaningful conversations and collaborations, emphasizing strategic collective action to address the societal impacts of AI. The forum underscored the critical role of philanthropy in advancing social equity in the AI space and by continuing to prioritize inclusivity and collaboration PAI has helped foster a landscape reflective of our values and aspirations.
Philanthropy Steering Committee Members
Chan Zuckerberg Initiative
New America
Omidyar Network
Good Tech Advisory
Surdna Foundation
Google.org
Markkula Center for Applied Ethics
Data & Society
MetLife Foundation
Skoll Foundation
Surdna Foundation
AI Now Institute
Ford Foundation
Cohere For AI
Annie E. Casey Foundation
Ada Lovelace Institute
AI Collaborative
Amazon Web Services
PAI brought together the AI & Philanthropy Steering Committee and broader community of PAI Partners at theAI & Philanthropy Forum in March 2024. The forum fostered meaningful conversations and collaborations, emphasizing strategic collective action to address the societal impacts of AI. The forum underscored the critical role of philanthropy in advancing social equity in the AI space and by continuing to prioritize inclusivity and collaboration PAI has helped foster a landscape reflective of our values and aspirations.
Vice President of Learning & Impact
Surdna Foundation
Advancing Safety in AI
This year, PAI took proactive steps to address the deployment of AI systems in contexts where safety risks can have widespread consequences, including healthcare, finance, transportation, and media. Recognizing the urgent need for robust safety measures, we convened over 50 experts through a collaborative workshop with GitHub to explore safeguards for state-of-the-art open foundation models and roles and responsibilities within the AI value chain. We also published 3 pilot study reports on our ABOUT ML initiative, which promoted the standardization of AI/ML documentation to enhance transparency and foster societal trust within organizations. Through comprehensive engagement, PAI continues to lead in establishing the foundation for responsible AI.
An Inclusive Economic Future for AI
In 2024, PAI championed the dialogue on AI’s role in shaping an equitable economic future. As AI continues to redefine automation and wealth distribution, we united partner organizations, economists, and worker representatives to craft actionable steps to ensure AI developers and policymakers were equipped to handle the challenges AI poses. Our work yielded the Path for Developing Responsible AI Supply Chains, a five-step guide to enhance accountability and governance. We released resources for public comment such as the Vendor Engagement Guidance and Transparency Template, essential tools for fostering responsible practices and improving transparency within AI-driven data supply chains. These initiatives mark strides toward shared responsibility and greater accountability in the evolving landscape of AI and labor.
PAI in the Media
Beyond the Code BBC Collaboration
PAI developed Beyond the Code, a digital branded series produced by BBC StoryWorks that demystifies artificial intelligence for global audiences.
Media Appearances
2024 by the Numbers
that is international, inclusive, and equitable
about the social and societal impact of AI
by governments
in all sectors and broader communities
PAI Community
New Members of the Board of Directors
In 2024 PAI welcomed six new members to its Board of Directors. The new board directors include leaders from the philanthropic, technology, and financial sectors, as well as civil society and academia, bringing a wealth of knowledge and expertise in advancing the responsible and ethical development and use of AI.
Esha Bhandari
ACLU
Natasha Crampton
Microsoft
Vukosi Marivate
University of Pretoria
Lori McGlinchey
Ford Foundation
Premkumar Natarajan
Capital One
Suresh Venkatasubramanian
Brown University
Esha Bhandari
ACLU
Natasha Crampton
Microsoft
Vukosi Marivate
University of Pretoria
Lori McGlinchey
Ford Foundation
Prem Natarajan
ACLU
Suresh Venkatasubramanian
Brown University
17 New Partners joined in 2024
PAI welcomed 17 new partners to our global network of more than 100 organizations to identify AI’s most important challenges and opportunities and co-develop practical, effective guidance for responsible AI.
Who We Are
At the heart of our work is a global community of partners committed to the responsible development and use of AI. With the help of our PAI community we are able to uphold the tenets that guide us in advancing positive outcomes for people and society.
- We will seek to ensure that AI technologies benefit and empower as many people as possible.
- We will educate and listen to the public and actively engage stakeholders to seek their feedback on our focus, inform them of our work, and address their questions.
- We are committed to open research and dialogue on the ethical, social, economic, and legal implications of AI.
- We believe that AI research and development efforts need to be actively engaged with and accountable to a broad range of stakeholders.
- We will engage with and have representation from stakeholders in the business community to help ensure that domain-specific concerns and opportunities are understood and addressed.
- We will work to maximize the benefits and address the potential challenges of AI technologies.
- We believe that it is important for the operation of AI systems to be understandable and interpretable by people, for purposes of explaining the technology.
- We strive to create a culture of cooperation, trust, and openness among AI scientists and engineers to help us all better achieve these goals.
PAI’s Staff
Thanks to our Funders and Partners
PAI is an independent, nonprofit 501(c)(3) organization and funded by charitable contributions from philanthropy and corporate entities. Our 990 forms from previous years can be found on our funding page.
We thank all of our Partners who form this special community for their valuable insight and support.
