Skip To Main Content
Matthew Green on Muck Rack

Matthew Green

As seen in: The New York Times, Forbes, HuffPost, Nature, The Guardian, The Wall Street Journal, Slate, Flipboard, The Legal Intelligencer, Aleteia and
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work. Claim your profile

Get in touch with Matthew

Contact Matthew, search articles and posts on X, monitor coverage, and track replies from one place.

Learn more about Muck Rack

Actions

Share this page

Is this you?

As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile

Articles

前沿模型加密思维链被曝安全漏洞，弱模型可解码暴露商业机密与隐私

|
By Matthew Green
| Tiger Sniff
美国前沿模型厂商又暴露出一个重要安全漏洞。“攻击者”只需几步操作，就可能把模型刻意藏起来的思维链“扒”出来。这里面不仅有模型能力的秘密，还可能夹带用户隐私，甚至包含模型在处理有害问题时产生的危险信息。 比较直觉的解读是，这为“攻击性蒸馏”提供了一条成本更低的路径。但真正值得警惕的，可能不是蒸馏本身，而是这个漏洞暴露出的另一层问题：当安全性与产品设计、商业模式发生冲突时，前沿模型厂商究竟会把安全放在什么位置？ 论文披露的时间线尤其耐人寻味。密码学家、同时也是约翰斯·霍普金斯大学教授的Matthew Green，几个月前已经向OpenAI与Anthropic告知相关攻击方法，但两家公司都认为这种风险难以真正爆发；直到这支研究团队用实验把攻击完整跑通，相关漏洞才得到修复。 加密思维链，已经成为AI行业商业秘密与安全机制的交叉点。一旦这个秘密可以被弱模型重新解码，模型厂商失去的就不只是商业秘密，还有原本以为可以控制的安全边界，以及对模型内部行为的监督优势，当然还有它想防止的被蒸馏。 一步步沦陷...
Open in Who Shared Issue with article?

Some thoughts about Anthropic’s new cryptanalysis results

|
By Matthew Green
| blog.cryptographyengineering.com
Yesterday Anthropic published two new cryptanalysis results, both outputs of Claude Mythos, their (still) unreleased advanced model. The first of these results attacks a signature scheme called HAWK, while the second is an improved attack against reduced-round AES. Anthropic also released a blog post describing the research process that produced these results. A few people online have asked me what this all means.
Open in Who Shared View Duplicates (4) Issue with article?

Let’s talk about encrypted reasoning

|
By Matthew Green
| blog.cryptographyengineering.com
This is a quick post I wanted to write about a hobby project I spent a weekend on. It has little to do with real cryptography, and mostly doesn’t expose a particularly exciting vulnerability. But it did teach me a lot about frontier LLM APIs and coding agents. It also got me certified as an OpenAI “cyber researcher” which is something that doesn’t happen every day. In any case, please keep your expectations low. Who knows, perhaps someone else will find something exciting to do with this.
Open in Who Shared Issue with article?
See all 132 licensed articles (135 Total)