AI-Assisted Worker Documentation: Lessons from a Field Pilot in Informal Labor Markets
Webinar | Online
About the Event
Informal workers often rely on verbal agreements, leaving no record of work performed or wages owed. This limits both enforcement and evaluation, as disputes are common but difficult to verify.
This session shares lessons from a pilot of a WhatsApp-based tool that allows workers to document their daily work and access information on their rights. The tool uses speech-to-text and simple interfaces to reduce reporting burdens in low-literacy, high-noise environments.
We present early evidence on user engagement, data quality, and trust, and discuss the practical challenges of deploying AI-enabled tools in real-world settings. We examine what types of information workers are willing to record, how accurate and usable these records are, and how they may feed into dispute resolution processes.
The session focuses on what worked, what did not, and what this implies for scaling AI applications in evaluation. We also discuss how such tools can complement traditional survey and administrative data by generating high-frequency, worker-reported data in settings where formal records are absent.
This session shares lessons from a pilot of a WhatsApp-based tool that allows workers to document their daily work and access information on their rights. The tool uses speech-to-text and simple interfaces to reduce reporting burdens in low-literacy, high-noise environments.
We present early evidence on user engagement, data quality, and trust, and discuss the practical challenges of deploying AI-enabled tools in real-world settings. We examine what types of information workers are willing to record, how accurate and usable these records are, and how they may feed into dispute resolution processes.
The session focuses on what worked, what did not, and what this implies for scaling AI applications in evaluation. We also discuss how such tools can complement traditional survey and administrative data by generating high-frequency, worker-reported data in settings where formal records are absent.
Speakers
| 名称 | 标题 | Biography |
|---|---|---|
| Nikita Kohli | Postdoctoral Scholar with the Institute for Business & Social Impact at the Haas School of Business, University of California, Berkeley | I am a development and labor economist studying how broader economic factors shape employment outcomes in developing countries. My current research focuses on understanding the impact of unions in settings with high informality, and the barriers to job access for marginalized groups. |