{
  "id": 159,
  "url": "https://arxiv.org/abs/2607.06196v1",
  "title": "Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability",
  "summary": "Current AI safety evaluation and benchmarking frameworks predominantly rely on Western-centric culture-agnostic defaults that mask critical regional laws, socio-linguistic nuances, and cultural taboos, leaving Vision-Language Models (VLMs) vulnerable in global deployments. We introduce Pluralis v0.1: a novel multimodal, multi-regional, and multilingual dataset built from a culture-first perspective. Spanning 6,448 prompts across six Asia-Pacific countries (Bangladesh, India, Korea, Pakistan, Sin",
  "authors": "Alicia Parrish, Rajat Shinde, Sanket Badhe, Xinyi Bai, Sree Bhargavi Balija, Hua-Rong Chu et al.",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": "india",
  "published_at": "2026-07-07T12:21:52.000Z",
  "fetched_at": "2026-07-14T14:14:19.970Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/159",
  "original_url": "https://arxiv.org/abs/2607.06196v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}