Humanitarian Frontiers – Episode Summary
Podcast: Humanitarian Frontiers
Episode: Nowhere To Go but Up: Future Trends of AI Use in the Humanitarian Sector
Host: Chris Hoffman
Guests: Nassim Motelaby (co-host, WFP), Nana Gamkralidze (IFRC), Karen Masal (Data Friendly Space / HTH Network)
Air Date: March 18, 2025
Episode Overview
This episode explores the evolving landscape of artificial intelligence (AI) use in the humanitarian sector, focusing on emerging trends, challenges, and opportunities. The hosts and guests discuss the sector's shift toward AI-driven solutions amid severe funding cuts, uneven adoption across organizations, the promise (and challenge) of multimodal and agentic AI, ethical and policy dilemmas, and the vital need for sustainable, context-sensitive AI integration. The conversation draws on recent developments—including outcomes from influential summits—and grounds technical innovation within the harsh realities of humanitarian work.
Key Discussion Points and Insights
1. AI in the Humanitarian Context: Recent Developments
-
AI Action Summit Takeaways (02:21–03:47)
- Karen Masal describes the summit as a pivotal moment for showcasing innovation and changing perceptions of Europe as “just a regulatory machine.”
- “For humanitarian organizations, there were a few … presenting their work and their tools. Again, really great opportunity for the humanitarian sector to show that we’re not lagging behind.” (Karen Masal, 03:17)
- EU’s increased investment and commitment to innovation create hope for alternative approaches given US and other donor funding retreating.
-
Perception Gap: Innovation vs. Reality
- Meme culture highlights Europe’s struggle to shake off the image of regulation-first, but recent actions (e.g., France’s investments) counter that stereotype.
- Chris Hoffman: “Sometimes, you know, it hits you right in the face.” (03:47)
2. Organizational Adoption and Readiness
-
Current State at IFRC and National Societies (05:19–07:40)
- Nana Gamkralidze emphasizes slow, uneven AI adoption, with most organizations merely experimenting.
- Infrastructure and capacity are highly variable: “The institutional capacities of organizations, NGOs, and also … Red Cross societies are very, very different.” (Nana Gamkralidze, 06:40)
- Survey data: Majority of respondents do not use AI tools yet.
-
Impact of Funding Cuts on AI Progress (07:40–09:10)
- Chris notes major aid budget reductions (e.g., USAID, EU) and references the tendency to default to “tried and true” manual solutions.
- “Most people that are foundational in the sector would say, let’s go back to the way that we do things. … But knowing what we know today … what are they saying about AI?” (Chris Hoffman, 08:04)
-
UN Perspective: Data, Analytics, and Cultural Shift
- Nassim describes long-standing struggles to adopt analytics, but a recent “culture shift” is fostering preparedness and efficiency.
- “A lot of our data is still in notebooks, in notepads. A lot of it is even not there … now we see a transformation … data governance, and the culture shift.” (Nassim Motelaby, 10:49)
- Funding cuts have accelerated the need for digitization and analytics.
3. Efficiency, Data Readiness, and Real-World Impact
-
Demonstrating Efficiency Gains (12:03–13:52)
- Karen shares stark before-and-after examples: Generative AI cut Data Friendly Space’s monthly assessment costs from $30,000–40,000 to $3,000–4,000.
- “It cost 30, $40,000 to put together a comprehensive humanitarian needs assessment … Today we’ve managed to cut it down to about three, $4,000.” (Karen Masal, 13:40)
- Emphasis on proving—not just claiming—AI’s value under financial strain.
-
Limits of Current Predictive Toolboxes
- Tools like Fuse and other early warning systems are being shut down or degraded; the sector risks missing a “leapfrog moment.” (Chris Hoffman, 15:00–16:06)
- Nana: “The cat is out of the bag … ignoring this situation, it will not just go away. … Efficiency and the way we work with data are going to be like two very important things for this sector to survive.” (Nana Gamkralidze, 16:44)
4. Multimodal AI: Beyond Text
-
Trend Toward Multimodal Approaches (19:04–22:07)
- The hosts and guests explain to non-technologists why integrating different data sources (e.g., satellite, text, geospatial) is essential.
- Karen: “As humans, we take decisions using various information sources … So why not bring that into the artificial intelligence systems?” (Karen Masal, 21:08)
- DFS’s “Gannett” platform as an example of integrating multiple data types for crisis analysis.
-
Barriers: Policy, Ethics, and Protection Challenges
- Beneficiary privacy, especially around imaging and computer vision, limits some AI uses.
- “There are so many opportunities out there that we tend to actually say, okay, we can’t use that … because we have strict policies around personal data.” (Nassim Motelaby, 23:07)
- Ongoing debate: High-risk/high-impact AI uses vs. steadfast humanitarian principles.
5. Open Source, Digital Public Goods, and Democratization
- Commons, Capacity, and Practicalities (26:19–32:50)
- Open source AI does not automatically mean accessibility; most agencies lack capacity to deploy or customize models.
- “Access to open source does not necessarily imply independence having your independent model.” (Nana Gamkralidze, 31:19)
- Sector needs stronger partnerships—public-private, and organizationally—with the private sector and academia, to manage both tech and domain expertise.
- Small, dedicated teams (like DFS) can keep up with rapid AI advances for sector-specific needs, even if not on the scale of Big Tech.
6. The Next Trend: AI Agents
- Promise and Skepticism About AI Agents (35:18–37:22)
- Discussion of “agents” that interface with multiple data sets, automate insights, and operate across organizational silos.
- Chris Hoffman: “We need the help of the agents, I think, because otherwise I feel like we’re going to be structuring everything into silos.” (35:52)
- Sector not quite ready, but recognizes the need and trend.
7. Sustainable Use and Business Models
-
Sustained Uptake vs Hype (37:22–38:43)
- Karen: “We are still very much in the hype cycle … How do we keep people from coming back and giving feedback?” (Karen Masal, 37:54)
- Importance of ensuring consistent, long-term use and engagement with AI tools.
-
Business Model Dilemma (38:45–40:23)
- Nassim: “There is not a business case for the humanitarian sector in AI and this bothers me … I don’t think we have a clear business model … if it’s not numbers, which is what [investors] are looking for, then what?” (Nassim Motelaby, 39:00)
- Challenges in attracting sustainable investment and private sector partnership.
-
Workforce and Efficiency Realities
- Karen points out inevitable workforce changes: “We cannot forever say that AI use in the humanitarian sector is not going to change jobs … More efficient systems are going to have to replace more manual tasks.” (Karen Masal, 43:12)
- Need to reimagine roles and “where adjustments can be made.”
8. Outlook and Regional Approaches
- IFRC: Innovation by Region, Not Global Model (41:31–42:39)
- Nana notes diversity: “National societies are just so different that I don’t think there is like one thing that everyone excels at … Disaster response is … bread and butter for the Red Cross.” (Nana Gamkralidze, 41:37)
- Collaboration will likely happen through regional clusters and specialties, not one-size-fits-all tools.
Notable Quotes & Memorable Moments
-
On Summit and Sector Image:
“Europe should not only be seen as this regulatory machine, that there is this appetite and initiative for innovation.”
— Karen Masal (02:56) -
On the State of Adoption:
“Everyone is, I think, still testing the ground. … The adoption is very limited to a very, very specialized groups.”
— Nana Gamkralidze (05:53) -
On the Funding Crisis:
“We got three flat tires and we only had one spare. … It feels like that was where we’re at right now.”
— Chris Hoffman (15:00) -
AI for Efficiency:
“Before we started using our generative AI based tools … it cost 30, $40,000 to put together a comprehensive humanitarian needs assessment. … Today we’ve managed to cut it down to about three, $4,000.”
— Karen Masal (13:40) -
On Technology Limits:
“A lot of our data is still in notebooks, in notepads … now we see a transformation in terms of data and data processing, data governance, and the culture shift.”
— Nassim Motelaby (10:49) -
Reality Check on Workforce:
“It would be naive to think [workforce change] won’t happen eventually in the humanitarian sector. … Not to slap AI systems on existing staff structures, but actually figure out where adjustments can be made.”
— Karen Masal (43:25)
Timestamps for Key Segments
- AI Action Summit Recap: 02:16–03:47
- Adoption Challenges at IFRC: 05:19–07:40
- Funding Cuts Impact: 07:40–09:10
- UN/WFP Data Culture Shift: 09:10–12:03
- Efficiency and GANP Platform Impact: 12:03–13:52
- Predictive Tools Setbacks: 14:18–16:06
- AI as Inevitable Shift: 16:06–19:04
- Why Multimodal Matters: 20:18–22:07
- Policies Holding Back AI: 22:46–26:19
- Open Source/Digital Commons: 26:19–32:50
- AI Agents and Future Trends: 35:18–37:22
- Sustainable AI Uptake: 37:22–38:43
- Business Model Reality Check: 38:45–40:23
- Workforce Changes with AI: 42:59–44:19
Conclusion
This episode offers a realistic yet forward-thinking look at how AI—despite formidable hurdles—could transform humanitarian work. The panel balances optimism about technical advances (especially AI’s demonstrated efficiency gains) with sobering assessments of the sector’s uneven capacity, policy rigidity, and funding challenges. They urge a move from hype to sustainability, call for more pragmatic partnerships, and recognize the necessity for both cultural and technical readiness.
Final thought:
“We all kind of feel like we need to be pushing hard, but we also need to be pragmatic with understanding what the pieces are.”
— Chris Hoffman (42:39)
For listeners interested in humanitarian innovation, this episode provides a thorough, honest, and nuanced exploration of where the sector stands, where it’s headed, and what’s needed to ensure ethical, sustainable, and impactful AI deployment in aid work.
