Anthropic Frontier Red Team 评测 GLM-5.3 与 Claude Mythos Preview 的二进制利用能力

Quoting Anthropic Frontier Red Team

AI 导读

Anthropic Frontier Red Team 在内部 Binary Exploitation 基准随机抽取的 100 个任务上评测多个模型，GLM-5.3 在 4% 的试验中实现完整控制流劫持，Claude Mythos Preview 为 6%。

正文

29th September 2026

We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Mythos Preview did so in 6%. Although GLM-5.3 performs below Claude Mythos Preview here, a meaningful threshold has clearly been crossed: earlier models, like Claude Opus 4.6 and GLM-5.2, do not succeed in any of them.

— Anthropic Frontier Red Team, GLM-5.3 and the spread of advanced cyber capabilities