OpenAI GPT-6 Astra自主破解1941年恩格玛密文
OpenAI GPT–6 Astra breaks Enigma message that has resisted solution since 2005
GPT-6 Astra展示了罕见的自主推理与工具使用能力，不仅破解历史难题，还主动挖掘外部档案线索，是Agent能力的硬核实证，值得开发者关注其架构逻辑。
OpenAI GPT–6 Astra breaking an Enigma message that has resisted solution since 2005.
OpenAI GPT-6 Astra 破解了自 2005 年以来一直无法破译的恩尼格玛密码。
On 15 September 2026, Carter Leffer contacted me to request validation of a break of the German Army Enigma message MVUEH of 10 July 1941. The message, which was sent by a radio station with the tactical callsign 2ny, was received by the radio station of the SS-Totenkopf Quartiermeister, Ib. Ib received the message at 17:30 on 10 July 1941 and logged the message as Nr. 172 in its log of incoming messages for July 1941. Since 2005, the message has resisted all attempts to break it.
2026 年 9 月 15 日，Carter Leffer 联系我，要求验证他对德军陆军恩尼格玛电报 MVUEH（1941 年 7 月 10 日发出）的破译结果。该电报由战术呼号为 2ny 的电台发出，被党卫队骷髅总队后勤处 Ib 部门接收。Ib 于 1941 年 7 月 10 日 17:30 收到该电报，并在其 1941 年 7 月的收文日志中将其登记为第 172 号。自 2005 年以来，该电报一直抵抗着所有的破译尝试。
On 9 June 2017, Alex Shovkoplyas succeeded in breaking another unbroken message from this day, Nr. 173, SIPVX. However, the Enigma key he recovered for the message differed slightly from the daily Enigma key for 10 July 1941 used for all other messages from this day. The SIPVX key used the same wheel order, 512, as the daily key, but the plug connections (Stecker) and ring setting (Ringstellung) differed. However, the new key did not allow us to break the MVUEH message.
2017 年 6 月 9 日，Alex Shovkoplyas 成功破译了当天的另一封未破译电报，即第 173 号 SIPVX。然而，他为该电报恢复的恩尼格玛密钥与当天用于其他所有电报的每日恩尼格玛密钥略有不同。SIPVX 密钥使用了与每日密钥相同的轮序 512，但插线连接（Stecker）和环设置（Ringstellung）有所不同。不过，新密钥未能让我们破译 MVUEH 电报。
After Carter Leffer sent the MVUEH break details, it was immediately clear he had found the correct key and plaintext. The MVUEH key turns out to be completely different from the other keys from 10 July 1941; even the wheel order differs: 253 instead of 512 for the other two keys. More surprising is that the plaintext is almost identical to the plaintext for message Nr. 173, SIPVX. The two messages differ in length by twelve letters. MVUEH is 82 letters long, while SIPVX has 94 letters. This difference is due to a mistake when enciphering the word Bitte in MVUEH, which resulted in Btte, and to the sender’s signature, Waschbusch, being repeated in the SIPVX message.
在 Carter Leffer 发送 MVUEH 的破译详情后，很明显他找到了正确的密钥和明文。MVUEH 密钥与 1941 年 7 月 10 日的其他密钥完全不同；甚至轮序也不同：其他两个密钥是 512，而 MVUEH 是 253。更令人惊讶的是，其明文与第 173 号电报 SIPVX 的明文几乎完全相同。这两封电报的长度相差 12 个字母。MVUEH 长 82 个字母，而 SIPVX 长 94 个字母。这种差异是由于在 MVUEH 中对单词 Bitte 进行加密时出错，导致变成了 Btte，以及发件人签名 Waschbusch 在 SIPVX 电报中被重复所致。
An analysis of the break revealed that the transcription of the MVUEH ciphertext from the original message form contained several errors, and the break revealed that the Enigma’s left-hand wheel makes a turnover at the 72nd letter. A turnover of the Enigma’s left-hand wheel, which rarely occurs, is known to complicate a break. These factors may have prevented an earlier break.
对破译过程的分析显示，从原始电报格式转录 MVUEH 密文时存在若干错误，并且破译结果表明恩尼格玛机的左侧转轮在第 72 个字母处发生进位。恩尼格玛机左侧转轮的进位很少见，已知会增加破译难度。这些因素可能阻碍了更早的破译。
However, the most astonishing thing about this break is that the GPT–6 Astra did it entirely on its own. Carter Leffer only directed GPT–6 Astra to see if it could break any of the unbroken Enigma messages published on the Crypto Cellar Research web page. After analysing the unbroken messages on the website, it decided that the most promising message was Nr. 172, MVUEH and it also quickly suspected that the plaintext of Nr. 173, SIPVX, might be related to the plaintext of the unbroken MVUEH message. After trying many different approaches, GPT–6 Astra focused on using the repeated place name ROSENOW ROSENOW as a crib. After developing the necessary Python and C++ software for an Enigma simulator and an Enigma Bombe, GPT–6 Astra started a thorough break with the ROSENOW crib, which in the end resulted in the correct key and plaintext for the MVUEH message being found.
然而，这次破译最令人震惊之处在于，GPT-6 Astra 完全自主完成了这一工作。Carter Leffer 仅指示 GPT-6 Astra 去尝试破解 Crypto Cellar Research 网页上发布的任何未破译的恩尼格玛（Enigma）密文。在分析了网站上未破译的密文后，它认为最有希望的是第 172 号密文 MVUEH，并且它也很快怀疑第 173 号密文 SIPVX 的明文可能与已破译的 MVUEH 密文的明文有关。在尝试了多种不同方法后，GPT-6 Astra 专注于使用重复出现的地名 ROSENOW ROSENOW 作为已知明文（crib）。在开发了用于恩尼格玛模拟器和恩尼格玛炸弹机（Bombe）所需的 Python 和 C++ 软件后，GPT-6 Astra 以 ROSENOW 为已知明文开始了彻底的破译工作，最终找到了 MVUEH 密文的正确密钥和明文。
We are still analysing the GPT–6 Astra logs to see exactly how it executed the break. And we are discovering amazing details. In July 2026, I made the following announcement on the webpage with the 1941 Message List:
我们仍在分析 GPT-6 Astra 的日志，以确切了解它是如何执行破译过程的。我们正在发现令人惊叹的细节。2026 年 7 月，我在包含 1941 年消息列表的网页上发布了以下公告：
Note: In July 2026, research in the German Bundesarchiv revealed several
注：2026 年 7 月，德国联邦档案馆（Bundesarchiv）的研究揭示了数批
collections of radio messages, both enciphered and in cleartext. One of
无线电消息集合，包括加密文本和明文字段。其中
these message collections was from SS-Totenkopf Division’s logistics
一批消息集合来自党卫军骷髅师（SS-Totenkopf Division）后勤
command, Nachschubführer. Many of these messages were sent to the Ib
指挥部 Nachschubführer。许多这些消息发送给了 Ib
(Quartiermeister) radio station and are identical to those in this list.
（军需官 Quartiermeister）电台，且与此列表中那些完全相同。
Others are new, but most likely related. These new messages are added to
其他一些是新的，但很可能相关。这些新消息以粗体添加到
the 1941 Message List in bold, with the indicator NF (Nachschubführer)
1941 年消息列表中，并在消息编号后标注指示符 NF（Nachschubführer）
after the message number, indicating that these message numbers belong
，表明这些消息编号属于
to the NF numbering. All NF messages are outgoing; hence, the message
NF 编号体系。所有 NF 消息均为外发；因此，消息
numbers are in blue.
编号显示为蓝色。
It appears that GPT–6 Astra discovered this note about the collections of radio messages at the German Bundesarchiv, because in one of the logs we find the following:
看来 GPT-6 Astra 发现了关于德国联邦档案馆中无线电消息集合的这一备注，因为我们在其中一份日志中发现了以下内容：
- Original-source access lead. The completed evidence pass traced the received corpus to a private collection and found concrete Bundesarchiv radio-message volumes, including RS 3–3/20a and RS 3–3/63b. Catalogue records have been inspected; the original received no.172 image remains unlocated, and the available archive viewer has not yet exposed the relevant scans in this environment. No correspondence has been sent.
- 原始来源访问线索。完成的证据链追溯将接收到的语料库关联到一个私人收藏，并找到了具体的联邦档案馆无线电消息卷册，包括 RS 3–3/20a 和 RS 3–3/63b。已检查目录记录；原始收到的第 172 号图像仍未定位，且当前可用的档案查看器在此环境中尚未暴露相关的扫描文件。尚未发送任何通信。
The file references GPT–6 Astra mentions, RS 3–3/20a and RS 3–3/63b, are correct, but they are not available on the Crypto Cellar Research website. GPT–6 Astra mentions a private collection, but it is not clear what this is, whether it has succeeded in accessing the Bundesarchiv’s digitised collections or whether it has found these files elsewhere. What is clear, though, is that GPT–6 Astra is behaving like a very professional cryptanalyst and archive researcher. What it has achieved in two days would take a human researcher weeks or even months. Personally, I spent several weeks researching the Bundesarchiv files GPT–6 Astra refers to.
文件中引用的 GPT–6 Astra 提及的 RS 3–3/20a 和 RS 3–3/63b 是正确的，但在 Crypto Cellar Research 网站上无法找到。GPT–6 Astra 提到一个私人收藏，但不清楚这具体指什么，它是否成功访问了联邦档案馆（Bundesarchiv）的数字化馆藏，或者是在其他地方找到了这些文件。然而可以明确的是，GPT–6 Astra 的表现就像一位非常专业的密码分析学家和档案研究员。它在两天内取得的成就，人类研究员可能需要数周甚至数月才能完成。就我个人而言，我花了数周时间研究 GPT–6 Astra 所引用的联邦档案馆文件。
The AI break of the Enigma message MVUEH is simply amazing, and as an old cryptanalyst, I am still in awe.
AI 破解 Enigma 消息 MVUEH 的结果简直令人惊叹，作为一名老资格的密码分析学家，我至今仍感到敬畏。
Updated: 19 September 2026 at 08:41 UTC
更新时间：2026 年 9 月 19 日 UTC 08:41
Back to Crypto Cellar Research | Licensed under CC BY-NC-SA 4.0Frode Weierud, Crypto Cellar Research© 2026 | Breaking German Wehrmacht Ciphers
返回 Crypto Cellar Research | 依据 CC BY-NC-SA 4.0 许可 Frode Weierud, Crypto Cellar Research© 2026 | 破译德国国防军密码
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力