Seminar: Social Impact Evaluations of Frontier AI Models with Faisal Lalani

​AI development is moving fast — but who's shaping it? Not most of the people who will live with the consequences.

Faisal Lalani, Head of Global Partnerships at the Collective Intelligence Project, works on one of the harder problems in AI governance: how to channel public input into places where decisions get made. That means coordinating across civil society, domain experts, and governments worldwide — and wrestling with what democratic participation in AI looks like at scale.

In this session, Faisal will share findings from CIP's 2025 Global Dialogues Index Report — which surfaced perspectives from more than 6,000 people across 70 countries — and introduce weval.org, CIP's platform for running societal impact evaluations on frontier models. He'll dig into what gets lost when lived experience is stripped out of how we evaluate AI, and why standard approaches to public engagement so often fall short.

​The question underneath all of it: Can AI governance be democratic, and what would it take?