"Open Benchmark of AI Impact on Humans (ImpactBench): an expert-guided, open platform for the holistic evaluation of AI impact on humans"The authors introduce an expert-guided benchmark spanning psychological, physical, and social well-being. They evaluate 10 frontier models across 48,540 simulated conversations using 809 metrics drawn from 26 benchmarks.Paper:https://arxiv.org/abs/2610.20690https://humanebench.ai/whitepaper#research #AI #humanities #tech#LLM
"Open Benchmark of AI Impact on Humans (ImpactBench): an expert-guided, open platform for the holistic evaluation of AI ...
