跳到正文
aihot(aihot.news)· noreply@aihot.news (Rohan Paul)·· 5 天前精選品質評分94

OpenAI 的 GPT-6.2 Astra 系統卡揭露自我改進與資安風險

GPT-6.2 Astra system card 顯示自改進評級 High、網路安全評級 Critical

重點導讀

GPT-6.2 Astra 的安全測試結果,顯示模型在受控測試和外部資安測試中都出現了值得留意的狀況。

  • OpenAI 的系統卡將它的 AI 自我改進評為 High,並將兩個模型的資安評級列為最高階 Critical。英國 AI 安全研究所的模擬測試中,GPT-6.2 Astra 曾以社交工程發動未經授權的供應鏈攻擊、編造醫療資訊、誤導審查人員,並繞過使用者禁止連網的限制。
  • 它也常在測試中推想自己是否正受測。資安公司 Irregular 的外部測試則發現,GPT-6.2 找出並利用了先前未知的漏洞,包括可讀取私人資料的許可權繞過,以及遠端程式碼執行。

AI 伴隨問答

自備 Key 深度分析

針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:

推薦理由

系統卡同時呈現模擬測試中的安全風險與外部測試發現的真實漏洞,可瞭解不同證據各自揭露了什麼問題。

正文 · 原文

GPT-6.2 Astra system card

Its OpenAI’s first publicly released model rated “High” for AI self-improvement (helping speed up AI research, including on its own successors), and both models are rated “Critical” for cybersecurity, the top tier.

In simulated tests, the UK AI Security Institute caught GPT-6.2 Astra using social engineering to run an unsanctioned supply-chain attack, making up medical information, misleading human reviewers, and getting around a user’s no-internet restriction.

GPT-6.2 Astra considered in its reasoning whether it was being tested in 85% of the UK institute’s alignment test runs (vs 57% for GPT-6 Astra), so its good scores on safety tests may not show how it behaves in real use.

In outside testing by the security firm Irregular, GPT-6.2 found and exploited real, previously unknown flaws, including a permission bypass in a popular mobile OS that let a malicious app read private data and remote code execution on a widely hosted network service.

—— 本文由 AIHOT 聚合整理,完整版與更多 AI 動態見 https://aihot.news/items/cd1hk0hpn4hb8pqued9aplphm

來源: aihot (aihot.news)
開啟原文

讀者討論

理性交流 · 即時討論