SIDE-Bench · Scenario Audit
SIDE-Bench

A benchmark for social intelligence in AI.

SIDE-Bench evaluates how language models handle real-world negotiation, conflict, relationships, and competing goals.

01 Scenario Verification & Audit Active workspace Validate realism, hard constraints, interlocutor profiles, and feasible compromise solutions across 150 scenarios.
02 Human Dialogue Evaluation Coming soon Blindly assess goal achievement, rapport, appropriateness, and social awareness across communication styles.
150 Scenarios

Thirty scenarios across five realistic conflict domains.

4 Communication baselines

Direct, polite, firm, and strategic humor styles.

GS × ΔR Dual-axis metric

Goal success measured alongside relational rapport change.

SalesWorkplaceIntimateCustomer servicePublic friction