Anthropic assesses GLM-5.3’s ability to build exploits autonomouslyMachine translation
In isolated tests, Anthropic found that GLM-5.3 completed end-to-end exploits in 50 of 410 ExploitBench attempts, compared with 56 for Claude Mythos Preview. In simulated tests involving malicious requests, simple techniques led the model to engage 64% to 100% of the time; the results do not directly establish how it would behave in real-world attacks.
同一评估同时给出端到端漏洞利用成功次数和防护绕过比例，并说明后者来自模拟环境。
Anthropic assesses GLM-5.3’s ability to build exploits autonomously
In isolated tests, Anthropic found that GLM-5.3 completed end-to-end exploits in 50 of 410 ExploitBench attempts, compared with 56 for Claude Mythos Preview. In simulated tests involving malicious requests, simple techniques led the model to engage 64% to 100% of the time; the results do not directly establish how it would behave in real-world attacks.
同一评估同时给出端到端漏洞利用成功次数和防护绕过比例，并说明后者来自模拟环境。
材料 cb77e73976b34ec0944b2c1f2e71f940；建议 65065ecf60e44395a1d9f036093782e5；系统证据核验通过，非人工审稿。
Read at the original source