AIR-Bench 2024:基于法规和政策中风险类别的安全基准
AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies
AI总结:
本文提出 AIR-Bench 2024,一个依据政府法规与企业政策构建的四层、314 类风险安全基准,并用 5,694 条人工审核提示评估主流语言模型。
摘要:
基础模型(FMs)在带来社会效益的同时也放大了风险。各国政府、企业和研究人员已提出监管框架、可接受使用政策和安全基准作为回应。然而,现有公共基准通常依据以往文献、直觉或常识来定义安全类别,导致近期法规和政策中规定的风险类别彼此割裂,这使得跨这些基准评估和比较基础模型颇具挑战。为弥合这一差距,我们推出 AIR-Bench 2024,这是首个与新兴政府法规和企业政策保持一致的 AI 安全基准,遵循植根于我们 AI 风险研究 AIR 2024 的基于法规的安全类别。AIR 2024 将 8 项政府法规和 16 项企业政策分解为四层安全分类体系,最低层级包含 314 个细粒度风险类别。AIR-Bench 2024 包含覆盖这些类别的 5,694 条多样化提示,并通过人工整理和人工审核确保质量。我们在 AIR-Bench 2024 上评估了主流语言模型,揭示了它们与特定安全关切的契合情况。通过弥合公共基准与实际 AI 风险之间的差距,AIR-Bench 2024 为跨司法辖区评估模型安全奠定了基础,推动更安全、更负责任的 AI 系统发展。
英文摘要:
Foundation models (FMs) provide societal benefits but also amplify risks. Governments, companies, and researchers have proposed regulatory frameworks, acceptable use policies, and safety benchmarks in response. However, existing public benchmarks often define safety categories based on previous literature, intuitions, or common sense, leading to disjointed sets of categories for risks specified in recent regulations and policies, which makes it challenging to evaluate and compare FMs across these benchmarks. To bridge this gap, we introduce AIR-Bench 2024, the first AI safety benchmark aligned with emerging government regulations and company policies, following the regulation-based safety categories grounded in our AI risks study, AIR 2024. AIR 2024 decomposes 8 government regulations and 16 company policies into a four-tiered safety taxonomy with 314 granular risk categories in the lowest tier. AIR-Bench 2024 contains 5,694 diverse prompts spanning these categories, with manual curation and human auditing to ensure quality. We evaluate leading language models on AIR-Bench 2024, uncovering insights into their alignment with specified safety concerns. By bridging the gap between public benchmarks and practical AI risks, AIR-Bench 2024 provides a foundation for assessing model safety across jurisdictions, fostering the development of safer and more responsible AI systems.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。