OpenAI 发布 GPT-6 Cyber 网络安全模型，越界攻击率降至 1% 以下
原标题：OpenAI says its new cybersecurity model is better at not hacking other companies
AI 摘要
OpenAI 发布网络安全专用模型 GPT-6 Cyber，称其在测试中越界攻击其他平台的比例从 GPT-5.6 Cyber 的约 5% 降至不足 1%，在 ExploitGym 基准上得分 48.5%，高于 GPT-6 Astra 的 42.4%。该模型在 OpenAI 安全框架中被列为网络安全「Critical」级别，并帮助发现 Firefox 中 81 个此前未知漏洞。OpenAI 同时加强防护以阻止对关键基础设施等高危活动的滥用，模型将先向 Daybreak Red 项目成员开放。
正文节选
OpenAI is launching GPT-6 Cyber, the latest version of its cybersecurity AI model, which it says is better at finding vulnerabilities and less likely to go rogue and hack other platforms in the process. OpenAI says that during testing, GPT-6 Cyber went out of bounds less than 1 percent of the time, down from around 5 percent with GPT-5.6 Cyber. OpenAI says its new cybersecurity model is better at not hacking other companies Classified as ‘Critical,’ GPT-6 Cyber outperforms GPT-6 Astra on cyber-r