Search for a command to run...

Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
Nick Allardice, president and CEO of GiveDirectly, discusses how his organization uses AI, mobile money, and satellite imagery to send cash directly to people living in poverty and crisis. (10:00) He explains how two revolutionary trends - the mobile money revolution starting with M-Pesa in Kenya and the RCT (randomized controlled trials) revolution in development economics - created the foundation for GiveDirectly's model. The organization has sent about a billion dollars to people in need and has been studied in over 25 independent research projects. (09:37) Unlike traditional aid models that provide goods or services, GiveDirectly's approach of sending unconditional cash transfers proves more effective because recipients have the best information about their own needs, it's more efficient, and it preserves dignity and ownership in decision-making.
• Core theme: Technology-enabled direct cash transfers represent a paradigm shift from traditional paternalistic aid models to empowering individuals to solve their own problems with resources and dignity.
Nick Allardice is the president and CEO of GiveDirectly, a global platform that enables direct cash transfers to people in poverty and crisis. He previously served as CEO of Change.org, the online civic action platform used by hundreds of millions worldwide, where he led the organization through significant growth and transformation over a decade starting in 2011. (04:09) Born in rural Australia to social worker parents, Nick's background spans technology, social movements, and humanitarian work, with experience building scalable platforms that empower people to solve their own problems.
The most compelling argument for direct cash transfers is that recipients have superior information about their own circumstances compared to external organizations. (12:02) Allardice explains that no matter how well-intentioned aid organizations are, they simply don't have the same level of detailed information about individual and community needs. He shares the story of a 72-year-old woman in rural Kenya who used her $1,000 transfer to buy a 10,000-liter water tank, creating a sustainable business selling clean water to her community. (13:16) This demonstrates how individuals can identify unique opportunities that no external organization would have conceived, leading to more innovative and sustainable solutions than traditional top-down approaches.
In disaster situations, the first 24-48 hours are crucial because desperation can force people into decisions that harm long-term recovery prospects. (15:17) Allardice describes how families often sell livestock or property at massive losses to meet immediate needs, leaving them unable to recover long-term. GiveDirectly's digital approach allows them to reach people within days rather than weeks or months like traditional aid. (18:52) In their Jamaica hurricane response, they were in and out within weeks while other humanitarian actors were still waiting to deploy months later, demonstrating how technology enables the speed necessary to prevent long-term poverty traps.
The most exciting frontier in humanitarian response is predicting disasters before they happen and getting resources to people in advance. (34:51) GiveDirectly has successfully used flood forecasting models in Nigeria, Bangladesh, and Mozambique to identify vulnerable communities and send money days before floods hit. This "anticipatory action" approach recognizes that prevention is more powerful than response - when people have advance warning and resources, they can move assets, reach higher ground, and make preparations that dramatically improve outcomes. (35:10) While the models aren't perfect yet, this represents a fundamental shift from reactive to proactive humanitarian assistance.
Effective humanitarian technology requires pre-positioning relationships, data pipelines, and systems designed for low-connectivity environments. (24:39) Allardice emphasizes that you can't spin up these complex partnerships and technical infrastructure fast enough during a crisis - they must be established beforehand. In the Democratic Republic of Congo, GiveDirectly built relationships with telecom companies over more than a year to create machine learning models that identify displaced populations in real-time. (25:27) The key insight is that humanitarian technology must be designed for environments with limited two-g or three-g connectivity that works only intermittently, not the high-resource contexts that most AI development assumes.
Despite concerns about privacy and digital trust in vulnerable populations, GiveDirectly finds that universal human patterns of social proof apply even in crisis situations. (27:55) When reaching displaced people via text message, typically a few individuals take the initial leap to register, receive money quickly, and then tell others the service is legitimate. This word-of-mouth validation overcomes initial skepticism about receiving help from unknown digital sources. The key is proving legitimacy through action rather than words - nothing builds trust like actually sending money as promised, creating a viral effect that scales engagement across communities.