arXiv Computation and Language
African Languages AI Safety Readiness Report
Read the original on arXiv Computation and Language →The Flow has not summarised this story yet — read it at arXiv Computation and Language.
Related stories
arXiv Computation and Language
3d ago
AfriGuard: Safety and Robust Refusal For African Languages
arXiv:2610.26474v2 Announce Type: replace Abstract: Large language models (LLMs) designed to interact with users are typically trained to respond appropriately to harmful or toxic content. However, e...
arXiv Machine Learning
Aug 17
Language-Specific Gaps in AI Safety Training Datasets
arXiv:2608. 13695v1 Announce Type: cross Abstract: Large language model providers routinely cite multilingual safety benchmarks spanning a dozen or more languages as evidence that their models are safe for non-English-speaking users.
arXiv AI
Aug 18
LLM Safety Alignment in Low-Resource Languages: A Systematic Literature Review
arXiv:2608. 14626v1 Announce Type: cross Abstract: Large Language Models (LLMs) have achieved substantial progress in safety alignment, yet their safety guarantees remain significantly weaker in low-resource and multilingual settings than in high-resource languages.
Hugging Face Trending Papers
Aug 11
The Illusion of Cross-Lingual Safety in Low-Resource Languages
Safety alignment in large language models (LLMs) is largely developed in English, assuming these safeguards generalize across multilingual settings. However, this assumption remains underexplored and exposes a vulnerability in low-resource languages.
arXiv AI
Jun 2
TukaBench: A Culturally Grounded Jailbreak Benchmark for African Languages
arXiv:2606. 01322v1 Announce Type: cross Abstract: Safety evaluation of Large Language Models (LLMs) remains heavily English-centric, leaving Low-Resource Languages (LRLs), particularly African ones, critically underexplored.
arXiv Computation and Language
6d ago
Artificial intelligence is creating a new global linguistic hierarchy
The paper analyzes the distribution of artificial intelligence resources across 6,003 languages, revealing that a small number of languages dominate and that disparities are widening exponentially. It compares the diffusion of AI with earlier information technologies, noting a hype-driven spread pattern. The authors introduce the Language AI Readiness Index (EQUATE) to map technological, socio-economic, and infrastructural prerequisites for AI deployment, aiming to guide equitable diffusion of language AI.