Table of Contents
The panel brought together diverse perspectives from government, academia, civil society, and industry. A clear consensus emerged around one critical point: the current concentration of AI power in the hands of a few tech companies poses significant risks for society.
"The main divide is between AI builders and AI users," noted Clément Delangue, CEO of Hugging Face. "If we end up in a world where there are a few AI builders, mostly from Silicon Valley building for billions of AI users, I think we're in trouble." This sentiment resonated throughout the discussion, with speakers emphasizing the need to democratize AI development globally.
Vilas Dhar, President of the Patrick J. McGovern Foundation, pushed the conversation further by connecting AI development to fundamental human rights. "You cannot have a conversation about what AI empowers without rooting it in justice, without solving for dignity," he argued. Dhar called for a paradigm shift in how we think about AI governance, suggesting that decisions about AI's future shouldn't be left to technology company leaders alone.
The role of public investment emerged as a crucial theme. Amandeep Singh Gill, UN Under-Secretary-General, said: "There is no public interest in technology without public investments in technology." This includes investments in talent, universities, compute resources, and digital public infrastructure.
The panel also grappled with the tension between innovation and accountability. Dr. Abeba Birhane from Trinity College Dublin challenged the notion that more AI is always better, suggesting that sometimes what's needed is to "scale back and question what is the purpose of this AI." She emphasized the importance of criticism and scrutiny in developing AI systems that truly serve public needs.
Janet Haven, Executive Director of Data & Society, highlighted the dual nature of empowerment in the AI age: "It is disempowering not to have access to technology. It is also disempowering when a technology violates your fundamental rights." She called for robust governance frameworks that protect rights while enabling innovation.
The discussion revealed strong support for open-source AI as a path forward. This approach, championed by Hugging Face and others, was seen as crucial for democratizing AI development and ensuring transparency. The panel suggested that open-source initiatives could help bridge the gap between AI builders and users, enabling more diverse participation in AI development globally.
Prime Minister Petteri Orpo of Finland provided a government perspective, highlighting his country's commitment to open AI development and the importance of public-private collaboration. Finland's approach, including hosting one of seven European AI factories, was presented as a model for fostering innovation while maintaining public interest.
The panel concluded with a call to action for more distributed power structures in AI development. Rather than viewing AI through the lens of competition, speakers advocated for a collaborative approach that emphasizes shared resources, open innovation, and public good.
As AI continues to reshape society, the message was clear: ensuring AI serves the public interest requires deliberate action to democratize its development, robust governance frameworks, and sustained public investment. The future of AI shouldn't be determined by a select few but should reflect the diverse needs and values of global society.
"We need to stop thinking about the decisions made about AI as something that technology company leaders do," Dhar said. "We have to take back the control of those decisions."
