The challenge of meaningful product metrics is not unique to AI, but is something the sector has been grappling with ever since the mobile boom allowed us to reach actors in LMICs without physically engaging with them.

AI is now being used by research respondents to prepare, invent, or even answer entire interviews, raising deep questions about authenticity, agency, power, and the very integrity of qualitative research.

Generative AI‑powered chatbots in social and behaviour change programs are already nudging people toward action, but not without raising important questions about gender, quality control, and the very natural of digital evaluation.