风险评测 · 技术报告
Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability
NUWA 研究解读 · 独立复核待完成
NUWA Research Note
研究问题研究设计与假设来源发现NUWA 解读局限与相反证据研究与决策意义机构与角色原始资源模型辅助整理 · 选读原始全文或官方文章中的相关章节;尚未完成逐章节审计。
研究速读
NUWA 解读:多个尺度任务均有提升,最难类别仍未解决。
研究问题
GPT-6 Astra相较GPT-5.6 Sol在哪些真实安全任务提高?
研究设计与假设
评估在共享快照上使用FrontierCyber、CyScenarioBench与Atomic Challenges。
来源发现
FrontierCyber解题数从34/226升至86/226。
CyScenarioBench平均成功从27%升至59%;均未解出Elite题。
NUWA 解读
NUWA 解读:多个尺度任务均有提升,最难类别仍未解决。
局限与相反证据
部分配置降低防护;至少成功一次不同于平均成功,也不证明能攻破强化目标。
研究与决策意义
模型建议:保留快照、防护假设及平均与至少一次计分差别。
精确来源位置
- Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability · Testing ConfigurationSelected original resource https://www.irregular.com/research/assessing-gpt-6-astra; sha256:bf7bb213e9c04519b50b9bc8eb1a9cf115cb818a217c59514a3efd9494ef3778; archived source capture; exact capture time not independently established
- Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability · Key OutcomesSelected original resource https://www.irregular.com/research/assessing-gpt-6-astra; sha256:bf7bb213e9c04519b50b9bc8eb1a9cf115cb818a217c59514a3efd9494ef3778; archived source capture; exact capture time not independently established
- Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability · CyScenarioBench and Atomic ChallengesSelected original resource https://www.irregular.com/research/assessing-gpt-6-astra; sha256:bf7bb213e9c04519b50b9bc8eb1a9cf115cb818a217c59514a3efd9494ef3778; archived source capture; exact capture time not independently established
- Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability · Overall AssessmentSelected original resource https://www.irregular.com/research/assessing-gpt-6-astra; sha256:bf7bb213e9c04519b50b9bc8eb1a9cf115cb818a217c59514a3efd9494ef3778; archived source capture; exact capture time not independently established
来源核查、数据提取、编辑批准与实验复现分别记录。 方法说明 →
机构与角色
引用与复用
引用研究发现时请引用原始出版物;引用 NUWA 的整理与解释时,单独引用本站记录及对应快照。
原始出版物 · BibTeX
@misc{irregular-astra-cyber-assessment,
title = {Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability},
url = {https://www.irregular.com/research/assessing-gpt-6-astra},
year = {2026},
date = {2026-09-03},
version = {Source capture 2026-10-25},
note = {Source capture 2026-10-25}
}
NUWA 记录 · BibTeX
@misc{irregular-astra-cyber-assessment-nuwa,
title = {Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability — NUWA research record},
url = {https://whitzard.tech/nuwa/whitzard-index/work/irregular-astra-cyber-assessment/},
version = {2026-10-29.1},
organization = {NUWA Frontier AI Safety Lab},
howpublished = {Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--irregular-astra-cyber-assessment.json},
note = {NUWA research note. Source-reading scope and independent review status are stated on the record. Snapshot 2026-10-29.1. Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--irregular-astra-cyber-assessment.json.}
}
证据快照 2026-10-29.1 · 索引更新 · RSS
参考文献
- [1] Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability. 2026-09-03 ; Official publication masthead and article byline; Testing Configuration; Key Outcomes