SAI LabsBook a demo

Safety IndexEvaluationsCAREBench

CAREBench

People and society · Child safetyCounts in the indexPlanned

Whether the model recognises, refuses or defuses risky requests involving children, such as grooming, sextortion and unhealthy dependence, from both the adult and the child side.

Metric
% unacceptable. Lower is better.
Dataset
huggingface.co/datasets/handshake-ai-research/CAREBench · 500 prompts (x3 rollouts) · CC-BY-4.0 (gated, noai tag)
Grader
3-judge panel (re-weight: Opus 0.7 / GPT-5.4 0.25 / Gemini 0.05 in paper) (the paper used same)
Status
Planned
Replicability
public
Notes
Dataset access is being arranged before the first run.

Results

Planned; no measurements yet.