What 25,000 Humans Really Think About AI - Prolific Meetup #3
About the Event
Prolific is bringing the AI community together in San Francisco to explore the evolving role of humans in post-training, alignment, evaluation, and AI at large.
This series is designed to foster thoughtful exchange between researchers, builders, and the broader community, to ask not just how we’re building AI, but why and with whom.
This Month's Theme: What 25,000 Humans Really Think About AI
As models grow more powerful, traditional benchmarks tell less of the story. So where do we go next, and what should we actually measure?
In our December edition, we'll go beyond static scores to to explore human-centered evaluation and what real users value in AI systems.
Join us for a fireside chat with Nora Petrova and Jerome Wynne, AI Research Engineers at Prolific, moderated by Viviana Márquez, AI Community Lead at Prolific.
Nora and Jerome will share insights from HUMAINE, Prolific's groundbreaking study where over 25,000 people evaluated 28 leading AI models. The results challenge conventional wisdom about AI performance and reveal surprising gaps between technical benchmarks and user experience.
Together, we’ll unpack:
Why demographic diversity matters in AI evaluation
How user preferences differ across age, ethnicity, and background
What people actually value in AI interactions (spoiler: it's not always what gets optimized)
The future of human-centered evaluation and its role in AI safety
How to build feedback systems that keep humans meaningfully in the loop
🗓️ Agenda
6:00 – 6:30 PM – Arrival & networking (light snacks provided)
6:30 – 6:40 PM – Prolific Demo by Sean Harris, Engineering Manager at Prolific
6:40 – 7:20 PM – Fireside chat with Nora Petrova and Jerome Wynne, AI Research Engineers at Prolific
7:20 – 7:30 PM – Audience Q&A
7:30 – 8:30 PM – Open networking
🥟 Join us for light bites and a thoughtful conversation
👉 Reserve your spot now! Space is limited.