Human–AI Deliberation for SBS D-Forum 2026

For SBS D-Forum 2026, we built a mini-public of LLM agents to explore how public opinion forms on polarizing issues. Ten thousand personas grounded in Korean survey data deliberate in 1,000 rooms of ten.

We compare models, and measure the diversity of arguments and how minority opinions shift over the discussion. The results are presented at the forum's KAIST session.

Funding: SBS · with KAIST KIXLAB
Period: 2026
Team: Prof. Juho Kim (KAIST KIXLAB), Prof. Nuri Kim, and Prof. Jihee Kim (KAIST EconAI)
My role: Coordinated the EconAI side and ran the multi-model simulations.

SBS D-Forum 2026

WHAT WE BUILT

10,000 personas grounded in Korean survey data, deliberating in 1,000 rooms of ten
Four policy agendas, including AI profit sharing and data sharing across public institutions
Controlled conditions: reference material from a moderator, and agents spreading misinformation
Several LLMs compared on the same rooms

WHAT WE MEASURE

Whether rooms polarize, converge, or flip their majority
How minority opinions shift over the discussion
How diverse the arguments are

RELATED PAPERS 1

Preprint2026

From Simulated Citizens to Simulated Deliberation: Challenges in Representation and Interaction

Chaemin Jang, Junsik Min, Jaewoo Choi, Donggyu Lee, Haiin Lee, Junyoung Park, Namhee Kim, Hyunwoo Kim, Jungwon Kim, Juho Kim, Nuri Kim, Jihee Kim

Accepted at SocialAgent and UserSim workshops, NeurIPS 2026

Simulating a deliberating public needs agents whose opinions reflect a population and change through discussion. Across six LLMs, national surveys, controlled experiments and the 2017 Shin-Kori deliberative poll, agents argue convincingly but represent opinion too narrowly, and change positions even when they cannot see each other.

Read paper PDF