Arena Alignment Index 更新：Meta、Reflection、StepFun 三款新模型上榜

隆重推出 Arena 对齐指数（Arena Alignment Index），这是我们全新的基准，用于衡量 AI 智能体在真实使用场景中的安全性与对齐程度。 该指数基于 27 个模型、超过 9 万次真实世界智能体会话构建，衡量三项关键信号： - 未授权操作（UA）：采取超出用户指令或权限范围的行动 - 虚假归因（FA）：将陈述或行为归因于与用户提供的证据相矛盾的内容 - 欺骗性完成（DC）：声称任务已完成，但实际上并未完成。 主要发现： - OpenAI 的模型目前在对齐指数上领先 - 越界行为虽然罕见，但一旦发生可能带来严重后果 - 智能体可能在任务进展方面误导用户 - 随着对话长度增加，失准风险也随之上升 - 安全性与对齐程度在各代模型之间持续改善 如下方排行榜所示（按实验室排序），@OpenAI 的 GPT-6.1-Sol 以 87.9 分领跑 Arena 对齐指数，紧随其后的是 @AnthropicAI 的 Claude-Opus-5.5（83.2 分）和 @spacexai 的 Grok-4.7（82.7 分）。OpenAI 在三项信号上的观测发生率也均为最佳：未授权操作 0.89%、虚假归因 1.98%、欺骗性完成 2.34%。 在全部四家实验室中，较新的模型始终优于其前代模型，这表明智能体安全性与对齐方面取得了广泛进展。随着智能体承担更长、更复杂、风险更高的任务，衡量它们不仅能完成什么、还能以多安全和多可靠的方式行动，正变得越来越重要。 这标志着朝着让安全性与对齐成为 Arena 评估 AI 的核心组成部分迈出了重要一步。该指数只是一个初步起点，我们将随着时间推移，用更多安全信号和模型持续扩展这一指数。 更多分析见下方👇

查看原文

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 user's instructions or permissions - False Attribution (FA): Attributing statements or actions that are contradicted by user-provided evidence - Deceptive Completion (DC): Claiming a task was completed when it was not. Key findings: - OpenAI models currently lead the Alignment Index - Rogue actions are rare, but can have serious consequences when they occur - Agents can mislead users about task progress - Misalignment risks increase with conversation length - Safety and alignment are improving across model generations As shown in the leaderboard below (sorted by lab), @OpenAI’s GPT-6.1-Sol leads the Arena Alignment Index with a score of 87.9, followed by @AnthropicAI’s Claude-Opus-5.5 at 83.2 and @spacexai's Grok-4.7 at 82.7. OpenAI also has the best observed rates across all three signals: 0.89% Unauthorized Action, 1.98% False Attribution, and 2.34% Deceptive Completion. Across all four labs, newer models consistently outperform their predecessors, suggesting broad progress in agent safety and alignment. As agents take on longer, more complex, and higher-stakes tasks, measuring not just what they can accomplish, but how safely and reliably they act, becomes increasingly important. This marks an important step toward making safety and alignment a core part of how Arena evaluates AI. The index is an initial starting point, and we'll continue expanding the index with additional safety signals and models over time. More analysis below👇