The world risks missing out on many of AI’s benefits—or seeing them distributed unequally—unless efforts to build the architecture for equal access are accelerated, warned leaders at the SDG Lounge.
The discussion followed the adoption of two major agreements: the Global Pact for the Future and the Global Digital Compact. These frameworks build on recommendations from the UN Secretary-General’s High-Level Advisory Body on Artificial Intelligence, whose report was released last week.
“Things are moving fast in the world of AI,” noted UN Deputy Secretary-General Amina J. Mohammed. “In 2015, when we agreed on the SDGs, AI wasn’t part of the conversation. Now, the challenge is how to keep pace and ensure we bring everyone with us—focusing especially on those furthest behind. This requires expertise and investments at the country level.”
Infrastructure, Connectivity, and Capacity
Brad Smith, Vice Chair and President of Microsoft and UN SDG Advocate, called the UN report a critical step toward making AI work for humanity. He stressed the importance of assessing countries’ capacities, investing in infrastructure, and establishing governance systems to guide AI’s development.
For Ruth Porat, President and Chief Investment Officer of Alphabet, connectivity remains foundational. She highlighted Google’s progress in language accessibility, with tools now supporting 246 languages—110 added in just six months—enabling 100 million more people to access services in their native languages. However, Ms Porat emphasized that connectivity alone is insufficient: training and capacity-building are equally vital for people to effectively use these tools.
Mr Smith underscored the historical inequities in distributing transformative technologies. “More than 100 years after electricity reached Manhattan, 700 million people still lack access. And 35 years after cellular service began in the global north, two billion people still don’t have cell phones. Will AI follow this trajectory, or can we avoid repeating history?” He also argued that bridging this gap requires joint public and private sector investments. While private companies provide resources, largescale infrastructure will depend on public sector support, including international financial institutions and sovereign wealth funds.
Inclusive Participation and Global Consensus
Carme Artigas Brugal, Co-Chair of the UN AI Advisory Body, emphasized the need for public participation in shaping AI’s future. “We, the citizens, must decide the limits of technology.
Preventing harm requires global consensus, inclusive participation, and representation. We need to act now.” She then stressed that education is one of AI’s most promising applications. “We can democratize access to education globally, but first we must address the 2.5 billion people still lacking internet connectivity.”
Anna Makanju, Vice President of Global Affairs at OpenAI, highlighted the need to ensure AI remains accessible. “This is, in many ways, the most advanced and accessible technology in history. The question is: how do we ensure that all people can access it?”
Both Ms Artigas Brugal and Ms Makanju stressed the importance of addressing AI’s biases and flaws, particularly concerning gender inequality. They called for more inclusive AI development processes that incorporate diverse perspectives from the outset and equip women and girls with the tools and skills needed to thrive in an AI-driven economy.
From Vision to Action
Anne Bouverot, Special Envoy on AI for France, emphasized the urgency of translating global agreements into concrete actions. While the Global Digital Compact and the Global Pact for the Future provide universal guidelines, the next step is practical, targeted implementation to deliver tangible results.
Building a fair and inclusive AI framework represents a significant opportunity to create a better future for all. The global agreements are a strong start, but collaboration and action are now critical. With the right efforts, AI can drive real progress, create opportunities, and bridge global divides. The challenge is vast, but so is the opportunity to ensure AI works for everyone.
WATCH FULL SESSION HERE