快讯
Arena.ai发布Arena Alignment Index基准：基于9万+真实Agent会话评估AI安全对齐
Arena.ai推出Arena Alignment Index基准，用于衡量AI Agent在现实使用中的安全与对齐表现，数据来自9万多个真实Agent会话、覆盖27个模型，指标包括未授权行动（UA）等信号。转发者Boris Power评论称模型正快速变得更加对齐。
事件来源
查看原文
RT Boris Power Models are getting a lot more aligned, quickly! Arena.ai: Introducing the Arena Alignment Index, our new benchmark measuring safety and alignment of AI agents in real-world use. Built from 90K+ real-world agent sessions across 27 models, the index measures three critical signals: - Unauthorized Action (UA): Taking actions beyond the