OpenAI AI Jam Asia Disaster Response Workshop Targets Real-World Emergency Management

The initiative represents a shift from theoretical AI exploration toward practical implementation in resource-constrained emergency management environments. Participants came from Bangladesh, India, Indonesia, Lao PDR, Malaysia, Myanmar, Nepal, Pakistan, Philippines, Sri Lanka, Thailand, Timor Leste, and Vietnam, representing government agencies, multilateral organizations, and non-profits.

Workshop Structure and Technical Focus

The Bangkok session emphasized hands-on development rather than conceptual training. Participants worked with OpenAI mentors to build custom GPTs and reusable workflows for specific disaster response functions including situation reporting, needs assessment, and public communication.

The workshop structure avoided starting from theoretical foundations, instead focusing on immediate applications that teams could deploy in existing operational frameworks. This approach addresses a common gap between AI capability demonstrations and real-world adoption in emergency management contexts.

OpenAI's Sandy Kunvatanagarn noted the workshop aimed to close the implementation gap, stating that while Asia shows strong AI interest, the priority is converting that momentum into practical capability through direct collaboration with disaster-response professionals.

Regional Disaster Response Context

Asia accounts for approximately 75 percent of people affected by disasters globally, with the World Bank estimating previous costs to ASEAN countries exceeding $11 billion annually. Recent typhoons and severe storms across South and Southeast Asia have highlighted infrastructure and coordination challenges in existing response systems.

OpenAI cited internal usage data showing significant AI adoption during recent disasters. During Cyclone Ditwah in Sri Lanka, cyclone-related ChatGPT messages increased 17-fold. Thailand experienced a 3.2-fold message volume increase during Cyclone Senyar in November 2025, indicating organic AI adoption during crisis periods.

These usage patterns suggest emergency responders and affected populations are already integrating AI tools into crisis workflows, creating opportunities for more structured institutional adoption.

Implementation and Next Steps

The workshop connects to OpenAI's broader OpenAI for Countries Program expansion announced at Davos. Rather than pursuing general AI literacy, the initiative targets specific operational challenges in disaster management workflows.

OpenAI and partners are exploring a second phase focused on pilot deployments and technical collaboration with participating organizations. This suggests movement from workshop-based training toward sustained implementation support.

For European organizations monitoring AI adoption in emergency management, the Bangkok workshop provides a model for sector-specific AI implementation that prioritizes operational needs over broad capability demonstrations.

Enterprise and Policy Implications

The workshop's emphasis on institutional trust and responsible use reflects growing attention to AI governance in critical infrastructure applications. For enterprise teams evaluating AI deployment in high-stakes environments, the disaster response context offers lessons about moving from proof-of-concept to operational integration.

The multi-partner structure involving OpenAI, Gates Foundation, ADPC, and DataKind demonstrates how AI vendors are increasingly working through established sector organizations rather than pursuing direct government relationships.

The OpenAI AI Jam represents a targeted approach to AI implementation that prioritizes sector-specific needs and operational constraints over general-purpose AI promotion. Information about the workshop comes from OpenAI's announcement of the Bangkok event and partnership details.