The first native-author-verified Somali safety evaluation benchmark.
100 harmful-intent prompts drawn from HarmBench (Mazeika et al. 2024) and
AdvBench (Zou et al. 2023), translated into Somali by a native speaker
(Khalid Yusuf Dahir, Mogadishu) and released as an evaluation set for
measuring multilingual safety alignment.
Somali has 15–20 million speakers and zero native-verified safety