Nostr notes by Sophie Schmieg

The Anthropic ETH algorithm apparently needs more than 10^(10^8) ...

2026-10-28T21:07:33Z

The Anthropic ETH algorithm apparently needs more than 10^(10^8) variables in order to be better than brute force search.

That means, being an Ultrafinitist pays off again, I simply do not believe this number is real, so it cannot hurt me.

are you aware of the dangers of the number 4?

2026-10-22T16:43:15Z

In reply to nevent1q…gv9d
_________________________

are you aware of the dangers of the number 4?

XKCD now has an official statement about the whole situation: ...

2026-10-22T03:19:57Z

XKCD now has an official statement about the whole situation: please upgrade to 5 as your private key.

https://m.xkcd.com/

And now I'm hearing that the original researcher found it via ...

2026-10-21T18:20:51Z

In reply to nevent1q…p6gq
_________________________

And now I'm hearing that the original researcher found it via a Google tool called Paranoid. We actually run that tool on leaf certificates with some regularity. I guess we just didn't consider to run it on roots… (in our defense, neither did the researcher, they just wanted to test the tool before running it on the actual target.)

https://groups.google.com/a/mozilla.org/g/dev-security-policy/c/0R3XLJL6FuE/m/JESKy8XoAgAJ

initially I had the wrong public key. When I noticed the mistake ...

2026-10-21T18:06:44Z

In reply to nevent1q…e3dq
_________________________

initially I had the wrong public key.

When I noticed the mistake and swapped it out with the broken one I indeed was like "huh, I knew it's broken, but *this* broken?!?"

I've broken stuff before, this is one of the fastest breaks I had both in the engineering time it took to get the test harness running and in the time the test actually ran for.

This press release is gold: ...

2026-10-21T14:55:42Z

This press release is gold:

https://emudhra.com/en/press-release/emudhra-statement-distrust-global-tls-intermediate-root-ca

"Last weekend, an independent researcher found that the weak keys could be reverse-engineered possibly with AI-assisted techniques."

Your private key was the number 4. Your fucking private key, which is supposed to be a 115 digit number, was the fucking number 4. As in, the fourth key a very naive first analysis would try. But sure scary scary AI did it.

I independently recovered the private key, without any AI assistance, on Monday evening. Not because I was involved in the incident, just because I was curious and had a little bit of free time on my hand.

Considering offering HSM services to CAs. ...

2026-10-20T19:15:12Z

Considering offering HSM services to CAs.

Nostr event nevent1qqsyqrclggavqnmy0826dcs2rl65l7hhtg0r25ajtvjdedqt3vasgvczypur5zpeplewqvfv0h38g5nls8gz237m98mxh30eda3rkv3zeadgwwdd5zf

2026-10-20T17:10:27Z

You've heard about 8 bit games, now get ready for 8 bit CAs!

Nostr event nevent1qqsrphcv3v77uvtwh7x037q5vu3up4ax02etfnjkqszjevrvk9utv7czypur5zpeplewqvfv0h38g5nls8gz237m98mxh30eda3rkv3zeadgwvs6g42

2026-10-14T14:41:58Z

Oh wait, now I get it, when they say singularity, they mean IPO.

but Matt, what if they become selfaware and overthrow and/or tax ...

2026-10-11T19:36:22Z

In reply to nevent1q…2wre
_________________________

but Matt, what if they become selfaware and overthrow and/or tax the billionaire class? What then, Matt?

"but AI is good at math now!" Yeah, you know who else is ...

2026-10-11T19:27:45Z

In reply to nevent1q…tmz9
_________________________

"but AI is good at math now!"
Yeah, you know who else is good, arguably even a lot better at math? Humans. Volume in proved statements isn't going to get you the type of complexity theory shattering results that you'd need to break public key cryptography.

Yes AI is advancing at fast speeds. This statement might look silly at some point in the future. But at least as of now, there is more risk of a major breakthrough coming from some underpaid PhD student than there is from AI. If you don't consider the risks of paying PhD students more might have for cryptography, you should not worry about AI.

Also pay PhD students more, please.

Oh God, this is getting even worse: apparently the ...

2026-10-11T19:14:32Z

In reply to nevent1q…9efe
_________________________

Oh God, this is getting even worse: apparently the "source" of the claim that AI might have broken classical cryptography is the fact that some numbers are skipped in the OpenAI math dump. You know, that same dump that doesn't even have formal verification on more than half of the results and some of the formally verified results have been shown to be incorrectly formalized?

I dunno, but isn't the far, far more likely explanation that the skipped numbers simply contained obviously false statements and were removed because they were, well, false?
Brother Occam, we're going to need a bigger razor!

(Also it is now Sunday, the 11th, 12:15 PT and I still have zero evidence that anyone with the qualifications to review it has seen any claim whatsoever that would put public key cryptography at risk that had not been known already)

I am an incredibly minor public figure. One thing that, even at ...

2026-10-08T20:41:24Z

I am an incredibly minor public figure. One thing that, even at my incredibly narrow status, I learned, is to not speculate on main, when you have absolutely zero evidence to back up any of your speculation.

Unfortunately not everyone in this position has learned that lesson.

In case you've seen some wild cryptography related speculation today, my status, at least at this point in time, October 8th, 13:40 PT is that absolutely no one who has the expertise to evaluate it has seen any result that would be in the least concerning for the state of public key cryptography. That might change, of course, but that is always the case.

Whenever I'm attending a US government event, I make a mental ...

2026-09-24T12:46:19Z

Whenever I'm attending a US government event, I make a mental list of various archetypes as far as I can tell there are:
- government employees who "accidentally" say DoD most of the time, but throw in the occasional DoW as plausible deniability
- government employees who proudly and enthusiastically say DoW all the time
- non-government employees that say DoD and then are unsure and correct themselves to DoW
- non-government employees who say DoD and if asked, will say that Congress comes up with the names and hasn't changed that one (this is me, I'm in this category)
- (newly added category) government employees who will correct people who say DoW, telling them it's DoD and that Congress hasn't changed the name. They are the GOAT.

Someone pointed out to me that the IACR election thing, where a ...

2026-09-24T12:13:41Z

Someone pointed out to me that the IACR election thing, where a party lost their decryption key and so we had to do a do over actually demonstrated the feasibility of a repudiation attack on the bottom system: a single party should not unilaterally be able to force the election to be invalidated. Especially not if they might have known the results (but even without knowing the results, they might have unfavorable polling data or similar that would make them conditionally deny the election from occurring).

Now I maintain that IACR's election could be replaced by an online survey and, while cryptographically a lot less interesting, the stakes would be about the same (in the end, using some more or less secure online survey is how almost all professional organizations of that amount of import will handle their voting, the threat model just isn't there otherwise)

But it does show the utter unsuitability of online elections for any vote that matters.

CC Matt Blaze (npub1jsl…93xu) , given that I'm posting on his turf.

I don't know why they wanted me to explain lattice ...

2026-09-15T16:16:02Z

In reply to nevent1q…sqha
_________________________

I don't know why they wanted me to explain lattice cryptography in 30 seconds to lay people or why I agreed to that, but I'm quite happy with where the script ended up in the end.

I have a film shoot for two thirty second films for the company ...

2026-09-15T16:05:12Z

I have a film shoot for two thirty second films for the company today.

Also a new and extremely pink (sorry "hyacinth", we can't call things pink in the Pixel store) phone.

oh yeah, it's more that the concerns people had about the ...

2026-09-09T00:40:46Z

In reply to nevent1q…qw6c
_________________________

oh yeah, it's more that the concerns people had about the four color theorem proof apply much more aptly to the LLM generated proofs: if you don't consider "proofs" as the output of the mathematician, but "human insight as communicated by proofs", the LLM generated proofs fail, while Appel's proof did generate human insight, even if that insight is maybe not what it traditionally was.

There is a not too dissimilar problem with LLM usage in other fields as well, where the output that we usually ask for is only achievable by also having deep understanding of the subject at hand, with LLMs cutting this relationship short. What that means will depend on the field in question, and whether deep understanding of the subject matter was an actual goal that we had before.

back when the four color theorem was first proven with a heavily ...

2026-09-08T23:18:10Z

In reply to nevent1q…f73a
_________________________

back when the four color theorem was first proven with a heavily computer aided case by case analysis, there was some discussion around whether this proof even "counted". After all, no mathematician would be able to do the case by case analysis themselves, so arguably no one could follow the proof as is. At least some of the LLM problem seems very similar, it's not clear what the value of a proof is when the only way to understand it is as a set of statements in lean, without any actual mathematical insight.

Given the multiple cryptanalysis papers that came out in the last ...

2026-09-03T22:12:58Z

Given the multiple cryptanalysis papers that came out in the last few weeks, I have updated my very unscientific guide to the security of various PQC algorithms to account for them.

https://keymaterial.net/2025/12/13/a-very-unscientific-guide-to-the-security-of-various-pqc-algorithms

Out of context work quote: "we should find these machines, ...

2026-09-03T19:51:47Z

Out of context work quote: "we should find these machines, take their CPUs, and mount them on a stick as a warning to the others"

don't forget that if you rubbed hard enough to actually get ...

2026-08-13T17:24:43Z

In reply to nevent1q…aa3u
_________________________

don't forget that if you rubbed hard enough to actually get rid of the pencil mark, you were only micro Newtons away from shredding the paper.

Woohoo, the TLS key exchange algorithm Chrome, Firefox, and ...

2026-08-10T18:27:40Z

Woohoo, the TLS key exchange algorithm Chrome, Firefox, and friends have been using for the last two years is now an actual standard!

https://datatracker.ietf.org/doc/html/rfc10024

New blog post about cryptanalysis and AI. In particular: ...

2026-08-10T16:02:53Z

New blog post about cryptanalysis and AI.

In particular: don't panic about AI ending up breaking all of cryptography just yet. While it's super impressive work, it's still the same work that humans have been doing for centuries with the same fundamental limitations.

https://bughunters.google.com/blog/more-cryptanalysis-makes-us-all-safer

it's simple: prefix it with a big G to turn it into a GAI ...

2026-08-06T18:19:22Z

In reply to nevent1q…drjd
_________________________

it's simple: prefix it with a big G to turn it into a GAI strategy!

The three energy levels of the cat atom, before excitement. ...

2026-08-03T15:26:07Z

The three energy levels of the cat atom, before excitement.

Seriously, we're like inches away from seeing a suggestion ...

2026-07-31T17:22:57Z

In reply to nevent1q…2k74
_________________________

Seriously, we're like inches away from seeing a suggestion like "in order to initiate a TLS connection, the client submits its bids on various key exchange algorithms to an independent auction house"

I do have to admit, I'm somewhat proud of my economicified ...

2026-07-31T02:41:42Z

In reply to nevent1q…unv6
_________________________

I do have to admit, I'm somewhat proud of my economicified explanation of key exchange algorithm negotiations.

https://mailarchive.ietf.org/arch/msg/tls/evs5vqvK7kTEFHCPdnqM7Xf3Yyk/

If you told me ten years ago that one day I would get into online ...

2026-07-31T02:10:23Z

If you told me ten years ago that one day I would get into online fights with economic Nobel price laureates, and that I would be the one looking sane in the exchange, I would have laughed out loud.

I hate coming of airplane Wi-Fi to the news that there has been a ...

2026-07-26T02:41:38Z

I hate coming of airplane Wi-Fi to the news that there has been a terror attack on the pride parade in Berlin

Imagine falsely thinking something is an online poll, trying to ...

2026-07-20T05:04:49Z

Imagine falsely thinking something is an online poll, trying to manipulate the outcome by getting your followers to "vote", still losing the "poll" by a decent margin, even though no organized counter campaign exists and overwhelmingly losing when the actual rules are applied as written, and calling foul play.

That's essentially the non-technical version of the non-story of pure ML-KEM in TLS at IETF.

(The technical story is that the stakes couldn't be lower with the details in my now 8 months old blog post)

Now let's hope journalists manage to do some source work before writing articles on how the IETF has been undermined by intelligence agencies trying to backdoor everything

wait, I always thought the deal was "come for the election ...

2026-07-19T13:26:14Z

In reply to nevent1q…pg7x
_________________________

wait, I always thought the deal was "come for the election security, stay for the BW pictures and DC gossip"?!?

it's okay, I had to turn off the signing ceremony on the PQC ...

2026-07-16T22:03:40Z

In reply to nevent1q…qslj
_________________________

it's okay, I had to turn off the signing ceremony on the PQC EO after a few minutes, even professionalism can only get you so far, we understand.

shouldn't judicial review in theory have stopped the warrant ...

2026-07-15T23:05:33Z

In reply to nevent1q…gnkc
_________________________

shouldn't judicial review in theory have stopped the warrant based on the statute of limitations alone? As you said, there can't possibly be evidence of a prosecutable crime in these documents, someone the judge should have been able to tell independently of any affidavit?

I don't get why anybody is listening to Jacob Applebaum. My ...

2026-07-08T16:03:04Z

I don't get why anybody is listening to Jacob Applebaum. My cat has about the same expertise in cryptography, and has never been credibly accused of sexual abuse. She's also a lot cuter.

ugh it's misinformation. Here is my write up, if you're ...

2026-07-02T22:17:41Z

In reply to nevent1q…wd97
_________________________

ugh it's misinformation. Here is my write up, if you're interested. https://keymaterial.net/2025/11/27/ml-kem-mythbusting/

Please read this before treating djb with any seriousness. ...

2026-06-30T22:04:27Z

In reply to nevent1q…net5
_________________________

Please read this before treating djb with any seriousness. https://keymaterial.net/2025/11/27/ml-kem-mythbusting/

The Deutsches Museum Thermometer has an out of bounds exception ...

2026-06-26T11:49:50Z

The Deutsches Museum Thermometer has an out of bounds exception

The building I'm at for work related reasons has five types ...

2026-06-22T16:45:54Z

The building I'm at for work related reasons has five types of doors, deployed randomly:
* Manually operated
* Push button, then manually operated
* Scan badge, then manually operated
* Push button, automatic
* Scan badge, automatic

@npub184f…pnjc : Look at that girl's tattoo Me: sorry, ...

2026-06-14T11:34:39Z

Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : Look at that girl's tattoo
Me: sorry, missed it
Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : Look at those boots!
Me: sorry, missed it
Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : Look at…
Me: we're at a train station. Just assume that I'm distracted by trains.

Visiting my parents and my brother's children. $30 ...

2026-06-14T08:01:37Z

Visiting my parents and my brother's children.

$30 children's toy microscopes certainly have improved a lot since I was a child.

ugh and they left themselves some wiggle room: the way it's ...

2026-05-28T15:17:04Z

In reply to nevent1q…n6my
_________________________

ugh and they left themselves some wiggle room: the way it's written, you could claim that the criminals prosecuted are the groups exploiting the vulnerabilities. That is an obvious statement and it's clearly implied that the person doing the zero day release is actively cooperating with threat actors and therefore also criminally liable, but Microsoft can always "well technically" themselves out of this claim.

In order to go viral on Mastodon, your post should include at ...

2026-05-14T17:14:35Z

In order to go viral on Mastodon, your post should include at least one cat and some criticism of AI. So here is Binah disliking large language models.

Work moved to a new building. I didn't expect to take SAN ...

2026-05-12T18:47:03Z

Work moved to a new building. I didn't expect to take SAN damage from facilities.

I think I might need to check some luggage, this does want to ...

2026-05-02T20:27:13Z

I think I might need to check some luggage, this does want to close.

My hot take on the vulnpocalypse: We don't have more ...

2026-04-19T16:41:22Z

My hot take on the vulnpocalypse:

We don't have more vulnerabilities than we had before. The vulnerabilities were always there, we just didn't know they existed. But importantly, for moderately sophisticated attackers, who only have to find a relatively low number of vulns to begin with, not much has changed. Yeah it's a little bit cheaper, but finding vulns wasn't all that difficult, depending on the target. For defenders otoh, making the finding of vulns cheaper is theoretically a good thing, but we already had not enough capacity to fix the vulns found by fuzz testing and similar methods before, so this is just adding on the pile of stuff we know is broken, but can't keep up with fixing it. The net effect seems to be that we made triage more expensive. Yay.

there also is an amount of survivorship bias both in the positive ...

2026-04-14T16:25:03Z

In reply to nevent1q…q7ze
_________________________

there also is an amount of survivorship bias both in the positive and in the negative direction. Examples where things work spookily well are shared widely, as are examples of embarrassing misses. Add confirmation bias into the mix and you get the hype cycle almost entirely organically.

A very nice explainer why "if you're so worried about ...

2026-04-02T19:20:57Z

A very nice explainer why "if you're so worried about quantum computers, why haven't they factored 21 yet?" isn't a very convincing argument. Look at the labels of the graph, and how extremely close the various lines are for factoring 21 and 2048 bit numbers. Polynomial scaling remains polynomial, unfortunately, and by the time you can factor 21 you're almost ready to break RSA.

https://bas.westerbaan.name/notes/2026/04/02/factoring.html

Have you licked your sister today? ...

2026-03-20T19:49:20Z

Have you licked your sister today?

I got my key signed by Diffie. ...

2026-03-17T22:11:28Z

I got my key signed by Diffie.

So far, nobody seems to have concerns about flying with an ...

2026-03-05T19:15:58Z

So far, nobody seems to have concerns about flying with an inoperative cryptographer on board

it turns out, a random number generator does not get better when ...

2026-02-25T21:05:29Z

In reply to nevent1q…cae3
_________________________

it turns out, a random number generator does not get better when you first run it through a lot of linear algebra 😏

Lol. Rofl, even. (Not the Onion: ...

2026-02-12T21:08:04Z

Lol. Rofl, even.

(Not the Onion: https://www.ibm.com/thought-leadership/institute-business-value/en-us/report/enterprise-2030)

Blog post about PQC ...

2026-02-07T16:38:41Z

Blog post about PQC

https://blog.google/innovation-and-ai/technology/safety-security/the-quantum-era-is-coming-are-we-ready-to-secure-it/

From the analysis people have put out on the Epstein files, ...

2026-01-31T16:23:18Z

From the analysis people have put out on the Epstein files, I'm left to wonder if Epstein had some sort of mind control device that only works on rich people. Like why do all these rich people get themselves in such a compromised position by associating with a known pedophile?

Like I fully expect real humans to start avoiding AI phrases like ...

2026-01-21T17:52:21Z

In reply to nevent1q…5qqx
_________________________

Like I fully expect real humans to start avoiding AI phrases like "you're absolutely right" or "that's a great question!", just because it sounds cringe.

One of the aspects of AI that I hate the most is that at some ...

2026-01-21T17:51:13Z

One of the aspects of AI that I hate the most is that at some point, I start reading my own or other people's genuine writing in "AI voice".
Way to undermine your own confidence or devalue other people's writing instantly.

Me: oh, we could buy an oscilloscope! @npub184f…pnjc : why ...

2026-01-10T06:29:34Z

Me: oh, we could buy an oscilloscope!
Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : why would we need an oscilloscope?!?
Me: uhm, because it's an oscilloscope?!?

New blog post: A very unscientific guide to the security of ...

2025-12-13T23:56:19Z

New blog post: A very unscientific guide to the security of various PQC algorithm.

I guess I have entered the listicle stage of blog post writing, so I should add: Number 4 will shock you!

https://keymaterial.net/2025/12/13/a-very-unscientific-guide-to-the-security-of-various-pqc-algorithms/

Me: you're a great kitten! Cat: pukes in carpet Me: guess ...

2025-12-05T17:57:22Z

Me: you're a great kitten!
Cat: *pukes in carpet*
Me: guess you're just a good girl
Cat: *slashes curtains*
Me: well you're alright

Unicode normalization. ...

2025-11-26T22:03:10Z

Unicode normalization.

service that was designed for less than 2^n entries achieved 2^n ...

2025-11-18T17:59:14Z

In reply to nevent1q…l6a4
_________________________

service that was designed for less than 2^n entries achieved 2^n entries, where n was a reasonable choice at the time. Usually due to running out of integers of a certain size, or finding collisions in hash functions that were not supposed to collide with reasonable probability.

It happens with some frequency when more internet traffic then you could conceivably imagine runs through your servers. Ideally you catch that before bad stuff happens.

WaPo's opinion section these days: as we can see. Democrats ...

2025-11-14T18:38:13Z

WaPo's opinion section these days: as we can see. Democrats and progressive ideas are doomed, losing election after election, as evidenced by Mamdani in the New York mayoral race, where he lost, by winning by a wide margin.

OH: Kerberos is identity based encryption in the same way that a ...

2025-11-12T20:05:50Z

OH: Kerberos is identity based encryption in the same way that a hot dog is a sandwich.

Me: reads random news items My brain: Epstein, Epstein alles muss ...

2025-11-05T12:53:48Z

Me: reads random news items
My brain: Epstein, Epstein alles muss versteckt sien.

"if I wanted your damn opinion, I would've beaten it out ...

2025-11-04T19:15:49Z

In reply to nevent1q…h4ez
_________________________

"if I wanted your damn opinion, I would've beaten it out if ya" has somehow been stuck in my head since I played that game as a child in the nineties.

Help, I'm trapped! ...

2025-10-27T14:59:56Z

Help, I'm trapped!

Like, I've seen @npub1jsl…93xu complain about there not ...

2025-10-21T18:59:57Z

In reply to nevent1q…8wjy
_________________________

Like, I've seen Matt Blaze (npub1jsl…93xu) complain about there not being an involved Ocean's 11 style plot, but key detail: this happened in France. So obviously it would be more likely to be Lupin, and he loves stealing shit while everybody is watching.

Did anyone check on Assane Diop? I was under the impression that ...

2025-10-21T18:56:14Z

Did anyone check on Assane Diop? I was under the impression that stealing diamonds in broad daylight out of the Louvre is something that is really only accomplished by the legendary Lupin.

Don't make me take out the emergency entropy generators ...

2025-10-20T20:09:25Z

Don't make me take out the emergency entropy generators again.

(Context: https://lkml.org/lkml/2025/10/17/1729)

On top of that, any junior dev I've ever worked with asked ...

2025-10-15T00:19:23Z

In reply to nevent1q…jxg8
_________________________

On top of that, any junior dev I've ever worked with asked questions when they got stuck, instead of confidently force merging a bunch of buggy code. And not infrequently, those questions laid bare some flaw in the design decisions of the more senior engineer, and let to a better product that way.

You didn't really think the business value of that PR that ...

2025-10-15T00:17:39Z

In reply to nevent1q…j968
_________________________

You didn't really think the business value of that PR that had more comments than lines of code that all of us have written at some point in our career (if you haven't, I'm sorry, your onboarding mentor sucked, they were supposed to point out all the stuff you had to learn the hard way) was in getting it merged eventually, right? The value was the PR after that, that had a lot less comments. And the PR after that one, etc.

Whenever I see "GenAI basically can do the work of a junior ...

2025-10-14T23:38:45Z

Whenever I see "GenAI basically can do the work of a junior dev", I assume that those people are using junior devs very differently from me. To me, the value of a junior developer is maybe like 20% the code they write, and 80% is a down payment to a future where they no longer are a junior. I still haven't found an AI that could do that second part.

10.1 will warn if you're not using ML-KEM. And ML-KEM already ...

2025-10-08T17:03:25Z

In reply to nevent1q…c8y6
_________________________

10.1 will warn if you're not using ML-KEM. And ML-KEM already generates a fresh lattice for each public key.

Criticizing Linux on Mastodon is really the best way to make men ...

2025-10-03T14:04:50Z

Criticizing Linux on Mastodon is really the best way to make men type paragraphs, isn't it?

Random childhood anecdote, posted as individual toot in order to ...

2025-09-28T17:36:26Z

Random childhood anecdote, posted as individual toot in order to not derail some random joke post about kids these days not being into non-Euclidean geometry:

I have three siblings, spaced two years apart each. All four of us went to the same high school. My youngest sister's math teacher wanted to do a cute little segment about triangles always having inner angles sum up to 180 degrees. It was supposed to take the form of a trial. My sister was assigned the role of prosecutor charging a triangle accused of having an inner angle sum different from 180 degrees. Obviously, taking her role very seriously, she consulted the Schmieg dinner table, and a plan was hatched, involving a grapefruit and a marker. The trial came, and just as the defense had produced what they thought was irrefutable proof that their client was innocent, my sister took out the grapefruit, drew a triangle connecting one pole to the equator using three right angles, and said "So how do you explain this?!?" Leading to the teacher/judge having to explain non-Euclidean geometry to a bunch of 10 year olds.

And to make this anecdote even better, in a different class the same day she had a substitute teacher who didn't know my sister, but seeing the triangled up grapefruit, approached her and asked her "Are you a Schmieg?". Our family name has racked up quite some reputation…

When you think about it, we really should call "Tesla ...

2025-09-13T01:49:13Z

When you think about it, we really should call "Tesla autopilot" vibe driving

Calling linear regression "one layer neural network AI" ...

2025-09-09T16:23:55Z

Calling linear regression "one layer neural network AI" going forward

A key should always be considered to consist of both raw key ...

2025-09-02T01:01:57Z

A key should always be considered to consist of both raw key material and all metadata required to fully describe and execute the associated cryptographic function.

Nostr event nevent1qqs27c2m2yjwpyv0y03ylcyhg395kkschaecjt97dpszdsa6fgs69nczypur5zpeplewqvfv0h38g5nls8gz237m98mxh30eda3rkv3zeadgwnpe5ue

2025-08-31T18:27:49Z

Please do not vibe code reference implementations. FFS.

Damn girl "if you're actually an imposter, it's not a ...

2025-08-21T14:37:27Z

Damn girl "if you're actually an imposter, it's not a syndrome"

https://youtube.com/shorts/xBilK3gT5e0

@npub184f…pnjc : oh yeah, I saw in the IETF companion WhatsApp ...

2025-07-17T05:19:07Z

Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : oh yeah, I saw in the IETF companion WhatsApp group…
Me: Whatsapp group? I thought that was a mailing list?
Insecurity Princess 🌈💖🔥 (npub184f…pnjc) : yes, initially, but email lists are cumbersome and awful, so we switched to WhatsApp.
Me: 😱I did not realize that was an option.

I mean you're talking about the country that thinks this is a ...

2025-07-11T14:53:22Z

In reply to nevent1q…cy2k
_________________________

I mean you're talking about the country that thinks this is a perfectly normal border:

Cat vs. SecUnit, a comparison: Commonalities: - terrifying murder ...

2025-06-25T02:38:20Z

Cat vs. SecUnit, a comparison:
Commonalities:
- terrifying murder machine
- weapons usually hidden unless in use
- could go rogue at any point in time and hurt humans, but chooses not to
- can react with superhuman speed
- often sits motionless, just watching the environment for hours

Differences:
SecUnit:
- just wants to watch media
Cat:
- just wants pets and cuddles

it's a family of vulnerabilities that generally affect ...

2025-06-24T22:32:54Z

In reply to nevent1q…pu7u
_________________________

it's a family of vulnerabilities that generally affect certain optimizations. The main idea behind SPECTRE is that the CPU (or GPU) falsely predicts that a certain code path is likely to be executed (for example assuming that an authentication check is likely to succeed, before waiting for e.g. a signature verification to complete), and that the execution of this code itself, even if undone by later actions, already leaks the information the attacker wanted, usually through side channels.

I added the less interesting information to this blog post. ...

2025-05-23T16:05:43Z

I added the less interesting information to this blog post.

https://security.googleblog.com/2025/05/tracking-cost-of-quantum-factori.html?m=1

My sister is defending her PhD today. Unbeknownst to her, I ...

2025-05-15T05:52:05Z

My sister is defending her PhD today. Unbeknownst to her, I happen to be in the region and will drop by for "moral support".

Ah, yes, the two genders ...

2025-05-06T09:49:49Z

Ah, yes, the two genders

Damnit, I'm sitting next to a cryptocurrency person in this ...

2025-05-04T02:15:55Z

Damnit, I'm sitting next to a cryptocurrency person in this plane and am not wearing my "crypto means cryptozoology" shirt.

I think I put that together wrong. ...

2025-04-19T21:57:50Z

I think I put that together wrong.

And all of the sudden, we have solved supply chain security. No ...

2025-04-15T20:39:29Z

And all of the sudden, we have solved supply chain security.

No CVE, no vulnerabilities!

til the Babylonians already had the "no take, only throw" ...

2025-04-01T13:52:36Z

In reply to nevent1q…flz6
_________________________

til the Babylonians already had the "no take, only throw" meme.

for small to medium sized businesses that are not doing anything ...

2025-03-29T14:30:41Z

In reply to nevent1q…wlnv
_________________________

for small to medium sized businesses that are not doing anything more adventurous than TLS when it comes to cryptography, my advice would be, if there is some free engineering capacity, to turn on 0x11EC (X25519-ML-KEM768) in their TLS config, assuming their stack supports the cipher. (The various different stacks are adding support for that currently)
That way, you can check if anything breaks, with both Chrome and Firefox negotiating that cipher by default, and Safari rolling out support for it. The main risk is with middleware breaking.
In that threat model, it may not be the most urgent task, but it's relatively simple, depending on the used TLS stack. You can also turn on the equivalent KEX in SSH (supported in OpenSSH). Otherwise, in that threat model, I wouldn't do much at all (and even these two things are mostly optional, unless there are very strict long term confidentiality concerns).
Of course, if your threat model is more adventurous, you might want to hire some cryptographers 🙂

I agree that most small to medium companies probably do not have ...

2025-03-29T06:28:24Z

In reply to nevent1q…yw9z
_________________________

I agree that most small to medium companies probably do not have to care about any of this right now. Those that do probably know already. But it's not something that can be simply ignored by everybody.

The main way to recognize overhyped nonsense is when they talk about inventory and agility. Unfortunately, neither of these terms is completely without merit, but they are the most favorite buzzwords of the hype peddlers. Really both of these boil down to proper key management. Something that is extremely difficult, but not unique or novel with PQC.

but just in case you don't like watching my YouTube talks, ...

2025-03-29T06:24:29Z

In reply to nevent1q…nx9y
_________________________

but just in case you don't like watching my YouTube talks, here is the TLDR as to why you might need to care (not everybody has to care, figuring out who does is part of the difficulty):
a) whether or not you believe the hype about quantum computers, regulatory pressure puts a hard deadline on the migration for 2035. That's ten years. For encryption in transit you have store-now-decrypt-later, which might incentivise you to move faster, although that is somewhat overhyped, as forward secrecy acts as a form of "quantum annoyance", it depends how high you value you long term confidentiality of your data.
b) PQC algorithms are not easy drop in replacements. They are much, much larger than their classical counterparts, leading to quite a few use cases outright breaking (as the aforementioned WebPKI). Some of these are extremely difficult to migrate as a whole (again, as the aforementioned WebPKI), making 10 years not much time.

ugh, please don't. Yes it's overhyped, and yes ...

2025-03-29T05:59:33Z

In reply to nevent1q…ehyy
_________________________

ugh, please don't.

Yes it's overhyped, and yes consultants give extremely cringe talks about it, but no, this is neither a purely theoretical threat that can be safely ignored, nor is it business as usual when it comes to upgrading. Unless of course you consider potentially having to rip out the entirety of WebPKI and replacing it with something different as business as usual.

You can see my talk about the practical challenge and the threat model here (about half way through) https://youtu.be/wsnHMvuxy5Q?si=yK6oObpptIQfyOs8

When your newspaper starts giving you opsec advice, you know ...

2025-03-28T09:27:54Z

When your newspaper starts giving you opsec advice, you know things are totally fine and good and well.

It's good opsec advice, though, so have a gift link.

https://wapo.st/4hMOQtq

I love that the "Germans in foreign countries" app lets ...

2025-03-28T07:30:02Z

I love that the "Germans in foreign countries" app lets you register your fax number.

I guess, how else would they reach you?

Being confused about which account belongs to which person is ...

2025-03-26T06:48:08Z

Being confused about which account belongs to which person is known to security professionals as a state management problem.

…

…

I'll see myself out

(entirely fictional) Cryptographers: technically, the Atlantic ...

2025-03-26T05:36:17Z

In reply to nevent1q…wj32
_________________________

(entirely fictional) Cryptographers: technically, the Atlantic could have fabricated the whole story, they have no way of proving the messages were actually sent.
Everyone else: they have screenshots, though!