[ivory-search id="3463" title="Default Search Form"]
Search
Register here
Search
Type below to search:
Solutions stage
Keynote
In person
LeadersGoldDiscovery

A 100 pathways for AI diffusion by 2030: A call to action

    • Date
    9 July 2026
    Timeframe
    11:30 - 12:00 CEST
    Duration
    30 minutes
    • Days
      Hours
      Min
      Sec

    The Challenge 

    The world’s most capable AI systems are being built and deployed at speed, but they are not reaching the people who need them most. For most communities in the Global South,  the barrier to AI adoption is not a missing model. Models exist. The barrier is that proven solutions, once deployed in one context, do not travel. Every new adopter starts from scratch. 

    This is the AI diffusion problem.  A voice AI system that works in one country can take months to adapt for another, not because the technology is different, but because the knowledge of how to deploy it never travelled. Who to involve, what to build on, what to watch out for: that kind of practical know-how is rarely packaged in a way that others can use. 

    The evidence suggests that most of what blocks diffusion is not technical. Roughly 70 percent of the challenge is systemic: whether institutions are ready to commission AI, whether the workforce can absorb it, whether the right partners exist to deliver it. The technology itself is the smaller part of the problem, and increasingly the easier one to solve. 

    What is missing is a structured way to capture how successful AI deployments actually happen, and pass that knowledge on. Not as a case study. As something the next adopter can pick up and use. 

    The Response: 100 Diffusion Pathways by 2030 

    In February 2026, twelve founding organisations, including Anthropic, UNDP, the Gates Foundation, Google.org, and EkStep, made a shared commitment at the AI Impact Summit in New Delhi: to document and disseminate 100 reusable AI diffusion pathways across the Global South by 2030. The initiative has since been endorsed by over 89 countries. 

     A diffusion pathway is that structured mechanism. It packages the lived experience of a successful AI deployment into two things: a deployment playbook capturing who needed to be in the room, the critical design choices, and the institutional arrangements that made it work; and a toolkit of technical components, safety guardrails, and training materials that the next adopter does not need to rebuild. 

     Pathways are already being built and tested across voice AI, language access, agricultural advisory, data readiness, and government capacity building. One example is the Voice AI Diffusion Pathway under the Italy, India, and Kenya Trilateral Agreement, where proven multilingual voice AI infrastructure is being adapted for low-connectivity environments across agriculture and financial services. The 2026 goal is 20 documented pathways, with visibility already to 12 to 15. The campaign is on track. 

    Schedule
    Share this session with your network
    • Session starts in
      Days
      Hours
      Min
      Sec
    Add to Calendar
    Related stories

    When anyone can tell their story

    A few years ago, Sonya Dukhon, a film director and creative producer, showed her co-writer Anna Pimen a short teaser...
    Global Summit

    AI is everywhere. Value isn’t. Here’s why people-centered transformation can...

    The pilot trap  The AI investment cycle has hit a paradox. Many organizations have allocated budgets, leaders report high confidence,...
    Global Summit

    Unveiling AI art on Lake Geneva

    A public art exhibition featuring artificial intelligence (AI) is sparking curiosity and conversations along the lakefront of the international district...
    Global Summit

    Are you sure you want to remove this speaker?