徐行

AI 工程师｜AI Agent · AI 编程 · AI 量化。分享有趣的实测和开源项目（AI engineer building agents & AI quant tools）不构成投资建议。

[徐行 @xuxing_ai]

商战朴实无华

[OpenAI Newsroom @OpenAINewsroom]

We discovered a sophisticated recording device that was hidden inside one of our offices. We used our models to understand the device and where it had come from, identifying that a specific employee had planted it a few days earlier. We referred the matter to the FBI, and we’re grateful to the agents and prosecutors who moved quickly to make an arrest. The FBI��s investigation is ongoing and we will continue to support their work.

At this time, we have no indications of this individual inappropriately accessing or exfiltrating model weights, source code, or customer data. The employee was terminated.

We take protecting our technology, data, and people seriously, and as a matter of national importance. Our security program is designed to protect against a wide range of threat actors, from insider risks to cybercriminals to nation states, and this case is an example of how AI can help address these sophisticated security threats.

[徐行 @xuxing_ai]

It's open source: a Claude Code skill + the 3D template + the recorder.
https://t.co/4QUOLUBPsE

[徐行 @xuxing_ai]

I gave Claude a floor plan. It built a home I can walk through.
It read the walls, doors and rooms, furnished every room, modeled the furniture and baked the lighting in Blender, all by writing code.
Built with Claude Opus 5.5.

[徐行 @xuxing_ai]

@1ittlecup 甜菜，想起只有一个手机时，需要另个手机扫描的囧境

[徐行 @xuxing_ai]

机器学习，深度学习虽然理论知识学过，但是没有机会实践过。现在AI agent的加持下，AI驱动式炼丹。claude code 自动调研模型，自动采集数据集，自行标注数据，自行提升���度。唯一不好的地方，只有入门级1050TI太慢了，一次跑几个小时，风扇呼呼转。卡到其他的3D视频做不了，氪金。

[徐行 @xuxing_ai]

google隔一段时间封一下，真 AI时代的 BAIDU。
antigravity 全回复： Agent execution terminated due to error.

[徐行 @xuxing_ai]

白嫖阿里云失败，已欠费，已穿仓。

[徐行 @xuxing_ai]

驮-01，一台会变形的仓储机器人。
四条腿能走台阶，腿一收就变成四轮车，到了货架，背上的机械臂自己把箱子搬上来。

[徐行 @xuxing_ai]

驮-01，一台会变形的仓储机器人。
四条腿能走台阶，腿一收就变成四轮车，到了货架，背上的机械臂自己把箱子搬上来。

[徐行 @xuxing_ai]

9/ 我自己的做法：
放在交易所的币，不为这事提出来，加固冷钱包是交易所的功课。
日常小额钱包照常用。
��期不动的大额，有空再挪到没转出过的地址，先小额试转，不连夜搬。
Vitalik 说他迁移失误亏的钱，比被黑的还多。

[徐行 @xuxing_ai]

1/ Drake 着篇"地堡模式"长帖 ，中文圈转的基本只有一边。
另一边是密码学家：Coinbase 密码学负责人 Lindell 直接说"这就是 FUD 的定义"。Vitalik 也说别慌着搬钱包。
AI 做数学，到底能不能破你的钱包？要不要动，值得思考 🧵

[Justin Drake @drakefjustin]

Today I call upon the blockchain industry to calmly begin planning for "bunker mode". My personal recommendation is to set in motion a controlled mass migration of assets to fresh addresses, i.e. addresses whose pubkeys remain hidden behind a hash.

Holders, starting with large and sophisticated ones, should consider moving the bulk of their funds to addresses that have never signed a transaction. And when they do sign one, they should also move remaining funds to a new address (possibly generated from the same seed phrase).

Don't rush. While I believe there is cause for action a rushed migration would do more harm than good. Don't panic either. Moving assets to protected addresses is a simple, preventative step which does not require new cryptography or new wallets.

IMO it is now reasonable to brace for the possibility that ECDSA breaks before qday, in the worst case in months not years. By "break" I mean fast private key recovery (e.g. in one week) on available hardware (e.g. a large GPU cluster).

Recent days have been humbling for human mathematical intuition. Long-held, unquestioned hypotheses have fallen. This includes the n log(n) bound for integer multiplication and the 3SUM conjecture. In hindsight, May's unexpected disproof of the Erdős unit distance conjecture was our warning shot.

Yesterday's OpenAI drop made it clear that mathematical superintelligence is upon us. They say there are weeks where decades happen. We are about to live through weeks where centuries of mathematical progress happen. Could our magic 64-byte ECDSA signatures be too good to be true? Was it just security through obscurity all this time?

Elliptic curves feel especially vulnerable to superintelligence. Curves carry rich structure, with room for fancy tricks like Schoof, Frobenius, pairings. (By contrast, hashes are designed to minimise algebraic structure.)

Separately, as Ewin Tang can attest, an efficient quantum algorithm sometimes foreshadows an efficient classical one. We should be open to the possibility of a classical counterpart to Shor that breaks elliptic curves and RSA at once.

Also noteworthy is the striking under-representation of cryptographic breakthroughs among the 722 mathematical results OpenAI published. I've witnessed first-hand the US government censoring academic quantum cryptanalysis results. Backroom interventionism is my base case.

I urge large, sophisticated actors to lead by example. Project11's "risq list" (bitcoin-risq-list.projecteleven[.]com) is a great tracker of exposed BTC pubkeys. Binance, Bitbank, Robinhood, Bitfinex, and Tether have an opportunity to harden their cold storage. Next month I'll address institutions in London in a live Q&A (forum.ethereuminstitutional[.]org/london-2026).

Again, please do not rush. Wallets holding under 50 BTC enjoy partial cover from "Satoshi's shield", i.e. his 20K exposed addresses that hold 50 BTC each. Load-bearing signers like oracles and L2 security councils should consider rotating ECDSA pubkeys with every signed message and/or multi-signing with a hash-based schemes like SPHINCS.

Exiting bunker mode safely will require post-AI cryptography. My inclination is to go all-in on hash-based cryptography and avoid structured mathematical assumptions entirely, whether from curves, lattices, or isogenies. A single battle-tested hash (e.g. from the SHA or BLAKE families) yields plausible post-AI security.

The Ethereum roadmap on strawmap[.]org fully embraces hash-based cryptography with end-to-end formal verification as a response to the quantum threat. Those timelines must now be revisited and accelerated in light of mathematical superintelligence. I'll be pushing for maximum defensive acceleration.

[徐行 @xuxing_ai]

8/ 我的看法：未雨绸缪，但不必草木皆兵。
2022 年 NIST 的两个候选算法 SIKE、Rainbow，就被新数学在普通电脑上打穿过，风险方向是真的。
但眼下没有证据显示 ECDSA 被 AI 打出实际攻击。"几个月"是 Drake 的最坏判断，连 Vitalik 都没给出这种时间表。

[徐行 @xuxing_ai]

7/ 反方也很硬。
Lindell 说，AI 能证明难定理，不代表被假设为困难的问题变容易了；也没有证据显示椭圆曲线比哈希更脆弱，历史上被破得更多的反��是哈希（找到碰撞，和从地址反推公钥不是一回事）。
他的结论："这就是 FUD 的定义。"

[徐行 @xuxing_ai]

6/ 证据到哪一步了？
OpenAI 这批成果里没有一篇是密码学的。
但 AI 确实动过手：今年 7 月，Anthropic 的模型把格类签名候选 HAWK 一组参数的破解难度降了约 6700 万倍（2^64 → 2^38），Anthropic 说没有实际影响。我的理解是，它和已经标准化的 ML-DSA 不是同一道数学题。

[徐行 @xuxing_ai]

5/ 为什么挪到新地址有用？
常见的比特币地址（1、3、bc1q 开头）和以太坊普通账户，地址是公钥算出来的"指纹"。没转出过的地址，链上只有指纹，公钥没露面。
但 bc1p 开头的地址收款就露公钥，转出那一刻公钥也会亮出来。所以这是争取时间，不是一劳永逸。

[徐行 @xuxing_ai]

4/ 这和喊了十年的"量子威胁"不一样。
量子要等专门的硬件，只打椭圆曲线和 RSA，解药是换成格密码这类后量子算法。
AI 找数学捷径不用新硬件，什么时候找到没法预测，还可能连解药一起打。所以 Vitalik 最担心的恰恰是格。

[徐行 @xuxing_ai]

3/ 做数学为什么能威胁钱包？
钱包安全建立在"某道数学题很难"上，但没人证明过它一定难。
Vitalik 举的例子：因式分解就被人类自己找到过捷径，所以 RSA 密钥得用几百字节。他担心椭圆曲线、还有抗量子的新算法"格密码"里也藏着捷径，AI 比人先找到。

[徐行 @xuxing_ai]

2/ 起因是 OpenAI 10 月 6 日公开了内部模型做的 700 多篇数学成果。
以太坊基金会的 Justin Drake 随后发长帖，说 ECDSA（你每次点"确认交易"用的签名算法）最坏几个月内就可能被 AI 破，建议大户先把币挪到没签过交易的地址，同时说"别急，也别慌"。

[徐行 @xuxing_ai]

@zhamomo666 失踪人员回归

[徐行 @xuxing_ai]

Try it in your browser (all data is simulated): https://t.co/k7DSNdhBgN
Code: https://t.co/ijGXUL1Xm0

[徐行 @xuxing_ai]

Most supply chain dashboards are a wall of tables. This one plays like a strategy game.

A factory in China ships to 3 overseas warehouses via 4 ports. Click any site and the camera dives in: cranes load the ship, trucks queue for dock doors, orders get picked on local time.