Nick Sullivan

The CFRG is holding a last call for two documents on hybrid KEMs: a mechanism that lets you safely combine a post-quantum KEM and a classical algorithm into a single construction. mailarchive.ietf.org/arch/msg/cfr... Reply on-list if you agree this should be published or have technical concerns.

This last call includes two documents: First: a generic combiners document that defines a general framework for combining supported by peer-reviewed research. Two combiners are defined: one universal, one optimized for KEMs with specific, strong binding properties. www.ietf.org/archive/id/d...

Every lock needs a key. The hard part is not always designing the lock. Sometimes it is getting the key to the person who needs it. That is the problem I look at in the third post in my CDT series on Encrypted Client Hello. cdt.org/insights/dis...

ECH is one of the most important privacy improvements happening in the web platform. But if we want it to work for the people most likely to face network interference, configuration delivery has to be treated as a first-class part of the problem.

Every lock needs a key. The hard part is not always designing the lock. Sometimes it is getting the key to the person who needs it. That is the problem I look at in the third post in my CDT series on Encrypted Client Hello. cdt.org/insights/dis...

The natural question is whether ECH configurations need to be tied so tightly to DNS. DNS currently handles both distribution and authentication. If those functions can be separated, ECH configs could be distributed through other paths and still verified.

Every lock needs a key. The hard part is not always designing the lock. Sometimes it is getting the key to the person who needs it. That is the problem I look at in the third post in my CDT series on Encrypted Client Hello. cdt.org/insights/dis...

That matters for ECH because a censor does not need to break the TLS handshake if it can prevent the client from obtaining the ECH configuration in the first place. The failure happens before the privacy property turns on.

Every lock needs a key. The hard part is not always designing the lock. Sometimes it is getting the key to the person who needs it. That is the problem I look at in the third post in my CDT series on Encrypted Client Hello. cdt.org/insights/dis...

China is the clearest example. The Great Firewall has a long history of DNS injection and poisoning. Encrypted DNS changes the mechanics, but not the basic incentive to control the discovery path.

Every lock needs a key. The hard part is not always designing the lock. Sometimes it is getting the key to the person who needs it. That is the problem I look at in the third post in my CDT series on Encrypted Client Hello. cdt.org/insights/dis...

ECH is designed to close one of the last major metadata leaks in HTTPS: the website name exposed in the TLS ClientHello. But before a browser can encrypt that part of the handshake, it needs the server’s ECH configuration.

ECH exposed a hard truth about privacy technology: you can win at the protocol layer and still lose at the deployment layer. I wrote about it here: cdt.org/insights/do-...

That is why signed ECH config updates are interesting. The goal is not just "more crypto." The goal is to remove the deployment constraint that created a stable fingerprint in the first place.

ECH exposed a hard truth about privacy technology: you can win at the protocol layer and still lose at the deployment layer. I wrote about it here: cdt.org/insights/do-...

That meant protected traffic still had a cheap classification handle. In other words: ECH stopped sticking out at one layer, and started sticking out at another.

ECH exposed a hard truth about privacy technology: you can win at the protocol layer and still lose at the deployment layer. I wrote about it here: cdt.org/insights/do-...

GREASE did part of that job. It made ECH-shaped traffic common, so the syntax itself did not stand out.

ECH exposed a hard truth about privacy technology: you can win at the protocol layer and still lose at the deployment layer. I wrote about it here: cdt.org/insights/do-...

I’m happy to be joining the USENIX Security ’26 Enigma organizing committee this year, after having the chance to speak at Enigma three times. It has a long history as a home for early, practice-driven security ideas, often where work first gets aired before it’s fully polished or widely deployed.

AI coding is an earthquake for software security. Not a tremor. The kind that liquefies the ground beneath your feet. We're mid-shake and most people are still debating if it's real. 🔗 github.blog/news-insight...

Liquefaction is what happens when shaking meets saturated ground. The soil loses structure and behaves like liquid. Buildings sink. In software: unverified code + relentless velocity + strained review = a codebase that can't hold weight.

AI coding is an earthquake for software security. Not a tremor. The kind that liquefies the ground beneath your feet. We're mid-shake and most people are still debating if it's real. 🔗 github.blog/news-insight...

Same survey: 96% of devs don't fully trust AI output. But only 48% say they always verify before committing. That gap is where bugs live. That gap is where security dies. 🔗 www.sonarsource.com/company/pres...

AI coding is an earthquake for software security. Not a tremor. The kind that liquefies the ground beneath your feet. We're mid-shake and most people are still debating if it's real. 🔗 github.blog/news-insight...

AI isn't coming; it's already in the pipes. Over 1.1M public repos now depend on an LLM SDK. Almost 700K of those appeared in the last 12 months alone. +178% YoY. 🔗 github.blog/news-insight...

AI coding is an earthquake for software security. Not a tremor. The kind that liquefies the ground beneath your feet. We're mid-shake and most people are still debating if it's real. 🔗 github.blog/news-insight...

Software Heritage archived over 22 billion unique source files by end of 2024. That's just public code they could find. The real number is unknowable, and growing faster than anyone can track. 🔗 annex.softwareheritage.org/public/annua...

AI coding is an earthquake for software security. Not a tremor. The kind that liquefies the ground beneath your feet. We're mid-shake and most people are still debating if it's real. 🔗 github.blog/news-insight...