Matthew Green

I teach cryptography at Johns Hopkins. https://blog.cryptographyengineering.com (#matthew_d_green on the other site.)

About me: I teach cryptography at Johns Hopkins.

Public Key

npub14ga56j26f2t92m7vuhl6z5vrave6wsgzkw7hy4qrnspsnqqsxkvq0cwcqv

NIP-05 Address

matthew_d_green_at_ioc.exchange@momostr.pink

Profile Code

nprofile1qqs25w6df9dy49j4dlxwtlap2xp7kva8gypt80tj2specqcfsqgrtxqpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dssfjrf4

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub14ga56j26f2t92m7vuhl6z5vrave6wsgzkw7hy4qrnspsnqqsxkvq0cwcqv

Show more details

Last Notes

2026-05-06 02:14:06 UTC

There’s something ominous about the speed with which the entire world has marched to require identification on platforms and, as I expected, begin the process of banning anonymous VPNs.

2026-03-25 13:54:12 UTC

- reply

My big concern is that in public spaces you have to throttle or gate people accessing the thing. For broadcast (read) only that doesn’t matter but for writes I guess you’d need permissioning or throttling.

2026-03-24 23:51:52 UTC

- reply

I’m planning to prototype this next week just for fun, to see if it works. But curious if there have been any experiments in this model or if it’s a bad idea for known reasons. (I know iOS background BLE receiving is very very bad.)

2026-03-24 23:50:47 UTC

- reply

All of these projects seem to operate on a 1:1 PC/radio model. I think a better design would move the client logic to a PC/smartphone, but each radio device could support many clients — each with their own public key/identity. You’d have to use BLE advertisements to support many clients per radio.

2026-03-24 23:48:39 UTC

- reply

Boston. #meshcore
https://files.ioc.exchange/iocexchange/media_attachments/files/116/286/823/405/963/232/original/4cab56a78ee1b267.mp4

2026-03-24 23:47:18 UTC

- reply

I’m planning to code this up next week.

2026-03-24 23:46:55 UTC

- reply

oh never mind, I understand what you’re saying. What I want is that but bigger: I want a single radio to be able to support 100 simultaneous smartphone clients via BLE advertisements. For protests and natural disasters.

2026-03-24 23:43:59 UTC

- reply

I think that’s basically what this does. It’s a real MeshCore node with all the radio layers turned into stdio ports. A better design would define this explicitly as a HAL.

2026-03-23 18:19:52 UTC

- reply

Right now it handles simulations that involve collisions, delays and other common events (in a very simple model). Each node is a simplified Meshcore node running as a process. #meshcore

2026-03-23 18:18:44 UTC

- reply

Just to give some visuals. #meshcore
https://files.ioc.exchange/iocexchange/media_attachments/files/116/279/863/631/130/422/original/5d3c45beba42e013.mp4

2026-03-23 18:17:15 UTC

I’ve been working on a little simulator for MeshCore, written entirely using Claude Code. I was hoping some MeshCore devs might take a look and let me know if there are features they could use. #meshcore

2026-01-23 13:59:23 UTC

Microsoft is handing over Bitlocker keys to law enforcement. https://www.forbes.com/sites/thomasbrewster/2026/01/22/microsoft-gave-fbi-keys-to-unlock-bitlocker-encrypted-data/

2025-12-25 18:04:22 UTC

I was stupid enough to buy this new AppleCare One plan for a phone I bought my daughter. Now I learn this only covers the device if it’s connected to the same Apple ID (not family plan). Have to spend Christmas unwinding this and getting a refund, what a drag.

2025-11-13 14:11:58 UTC

- reply

PS I did a search to see if there are any public court documents talking about Private Relay, and the closest I found is one CSAM case involving a defendant who used Private Relay, but was caught due to logging into an Adobe service. https://insidelake.com/2023/08/11/geek-squad-manager-arrested-on-more-than-30-charges-involving-child-pornography/

2025-11-13 14:11:18 UTC

- reply

I’m fascinated by these services for a reason: this is the first time that major companies have taken a stand on adding IP privacy to their offerings. It’s as big a deal as the rise of end-to-end encryption or TLS, but everyone noticed those things. Nobody is talking about this.
Also, what makes IP hiding services feasible today is the economics of the thing. With peering agreements and the extremely low bandwidth costs big tech cos can access, using 50GB/mo of Private Relay probably costs less than $1 in ingress/egress charges. Apple charges $1-10.

2025-11-13 14:10:18 UTC

- reply

Anyway, the current usage for both systems is web browsing (although Google seems to have given up on that.) But companies are also using this to add privacy to their own internal services, presumably things like Find My and (maybe) Siri and Apple’s PCC or Google’s new private ML inference service. https://blog.google/technology/ai/google-private-ai-compute/

2025-11-13 14:01:19 UTC

- reply

That’s what they claim.

2025-11-13 13:58:29 UTC

- reply

It’s not just Apple. Google has their own system in the “Privacy Sandbox” that goes by the (less cool) moniker “IP protection”. (Come on Google, get a better marketing team.) It’s experimental and was linked to Chrome, although they seem to have deprecated that integration (but not the service.) https://privacysandbox.google.com/protections/ip-protection

2025-11-13 13:57:43 UTC

- reply

It’s hard to prove that this “two company” solution really provides strong privacy. But it provides enough privacy that Apple’s instructions for Law Enforcement page explicitly warns police that they can’t trace Private Relay addresses.
https://files.ioc.exchange/media_attachments/files/115/542/737/520/660/482/original/9b29e7f7bf1a5ce6.jpeg

2025-11-13 13:57:10 UTC

- reply

For example, Apple has iCloud Private Relay for Safari browsing. The first “hop” is a server run by Apple itself, so Apple sees IP addresses. But web connections get relayed to a second VPN provider that isn’t Apple: hence Apple can’t link source IPs to destination servers.

2025-11-13 13:56:46 UTC

One of the most interesting recent privacy developments is the deployment of big two-hop IP blinding VPNs by companies like Apple and Google. These systems are designed to ensure that even those companies can’t link web requests to IP addresses.

2025-10-29 16:32:08 UTC

Great article about how TEEs are providing much less security than folks believe they will. https://arstechnica.com/security/2025/10/new-physical-attacks-are-quickly-diluting-secure-enclave-defenses-from-nvidia-amd-and-intel/

2025-10-16 12:29:34 UTC

- reply

People get upset about a lot of things enough that regulators and legislators try to do things about them. Data breaches get a lot of attention, for example. If Google was leaking detailed web histories of all its users, you can bet Congress would be shouting. But Google has been amazingly good at protecting this info from becoming public.

2025-10-16 12:27:17 UTC

- reply

Is a terrible problem that we still can’t get anyone energized about.

2025-10-16 12:23:21 UTC

- reply

Which is kind of the same thing, right? People say stuff like “the microphone must be listening to me because I saw this ad,” and companies say “we don’t listen to your microphone.” And then folks educate themselves, stop living their lives and make adtech behavioral modeling into the most important issue in their political lives. Except mostly they don’t. They just find it creepy and go back to scrolling.

2025-10-16 12:14:50 UTC

- reply

I have concerns too.

2025-10-16 12:13:49 UTC

- reply

Try it. Go out and try to convince your non-technical friends that adtech tracking is a weapon in the hands of a malicious government, and convince them to move it high up enough in their priority stack that they displace other concerns to try to stop it. I’ve tried and people nod and tell me how annoying ads are, then they move on. (And frankly, based on past history, they’re being rational and I’m being alarmist.)

2025-10-16 12:11:34 UTC

- reply

What I mean here is that drawing a line from ad targeting tech to “1984-style Orwellian behavioral surveillance” is very easy for tech people to do, since the tech exists to do a lot of this. But convincing people that this is a real threat is very hard, because it’s hypothetical and (so far) the ad tech companies have never allowed their data to be used that way.

2025-10-16 12:09:39 UTC

- reply

We (tech people) understand that this capability is possible if ad targeting tech was repurposed that way, particularly with sophisticated AI. But good luck convincing people that targeted ads are an existential risk to their freedom, rather than an annoyance.

2025-10-16 12:07:45 UTC

- reply

They don’t view ad targeting as an existential risk. They dislike ads. Some people find them creepy. But very few people are worried that the government is going to subpoena Google to convert their ad targeting model into a social credit score or even present that data as evidence in a criminal case. Because (I believe) it’s never happened.

2025-10-15 15:26:32 UTC

- reply

And right now I’m seeing very little deep thought about this stuff. We’re putting more tech and legislative and policy energy into stablecoins and TikTok than we are into how the entire Internet will work over the next 50 years. That should change!

2025-10-15 15:26:22 UTC

- reply

This is the world we’re moving into. It’s one where every online interaction of any small commercial value will be tied to a precise human identity, and that’s the only thing that will make the Internet even remotely usable in the future. Questions of privacy and accountability in that world matter.
We already leak a lot of this data, but critically *relatively little of it has been centralized and exploited* in the ways end-users are afraid of (they pretend to care about ads, but they don’t.) If we’re moving into this new world we need to think hard about tech and policy.

2025-10-15 15:25:38 UTC

- reply

But some of the people working on this tech seem to think that age verification is what this tech is for. I ask a question like “how will you handle [some type of fraud]” and they say “oh that problem doesn’t matter for verifying ages.” And it shocks me.
You’re building the architecture for how humans will interact online for the next several decades, and you don’t understand what task you’re engaged in, and you’re not thinking hard about how to do this privately? You actually think it’s about keeping 15 year olds from seeing salacious TikToks?

2025-10-15 15:25:00 UTC

- reply

Google recently released a toolkit to enable anonymous credentials and web authentication using zero-knowledge. This will likely be in Android and hopefully Apple will adopt it, which is a good thing! I’m not complaining! This will enable people to prove age privately. https://blog.google/technology/safety-security/opening-up-zero-knowledge-proof-technology-to-promote-privacy-in-age-assurance/

2025-10-15 15:24:21 UTC

- reply

Most of that tech stuff is already well under way. Several states already let you bind your driver’s license to your phone, for purposes that are (at present) entirely unconvincing. The interesting “hard” problems are all around privacy and fraud-resistance.
https://files.ioc.exchange/media_attachments/files/115/378/871/105/750/695/original/5c93fbe260188791.jpeg

2025-10-15 15:23:49 UTC

- reply

To be really clear: it’s pretty obvious that the central (Internet communication) problem of our time is going to be determining whether the stranger you’re talking to (or delivering ads to) is a person or a bot. And every existing tech we have for doing this will fail.
So how do we do this? Presumably by tightly binding physical identity to your device and then proving possession (with some other bells and whistles). Not coincidentally that’s exactly what age verification is. Weird how corporate and gov’t priorities suddenly align, right?

2025-10-15 15:23:11 UTC

The “age verification” and the “human identification” problem are the same problem. It upsets me to be around people who think they’re working on the first, but don’t understand they’re actually working on the second.
https://files.ioc.exchange/media_attachments/files/115/378/866/559/171/366/original/3fea85c209944528.mp4

2025-10-15 10:45:50 UTC

Two great bits of academic attack work this week. The new one allows Android apps to see data displayed on the screen by other apps, including Google Authenticator codes. https://www.pixnapping.com/

2025-10-14 11:24:54 UTC

- reply

This work really helps to illustrate the role of public and academic security research. Yesterday there were people at satcos who knew these unencrypted links were dangerous, but weren’t empowered to do anything about it. Now they will be.

2025-10-14 01:23:09 UTC

- reply

Some genuinely weird corporate security speak here. These downlinks are broadcasting your WiFi data across the entire northern hemisphere. It is not equivalent to a coffee shop!
https://files.ioc.exchange/media_attachments/files/115/369/901/118/190/811/original/d53ce36d69feb391.png

2025-10-14 01:18:08 UTC

- reply

Here’s a good article about it in Wired: https://www.wired.com/story/satellites-are-leaking-the-worlds-secrets-calls-texts-military-and-corporate-data/

2025-10-14 01:16:58 UTC

This is amazing research by Nadia Heninger and her co-authors Wenyi Morty Zhang, Annie Dai, Keegan Ryan, Dave Levin and Aaron Schulman. TL;DR a huge number of satellite links over our heads are totally unencrypted. satcom.sysnet.ucsd.edu

2025-10-10 11:40:41 UTC

- reply

And from a pure privacy perspective, the problem there is that once you’ve demanded every encrypted app add a scanning component, then changing what you scan for is just a software update to be agreed upon quietly in some committee down the line.

2025-10-10 11:40:27 UTC

- reply

So what worries me come December is that we see a bunch of “moderated” proposals that preserve this central architectural change, but just use it for slightly less. That’s not going to save anyone.

2025-10-10 11:40:06 UTC

- reply

The problem with this idea is that most “moderated” versions don’t solve the central problem: you’re changing private, encrypted messengers to have a component that scans the plaintext of the message. This is what potential makes your messages vulnerable to theft.

2025-10-10 11:39:47 UTC

Germany has agreed to stop ChatControl for now, due to huge amounts of public pressure. Good job! The bad news is that it could come back as soon as December, and the German government has interpreted the feedback as a need to “moderate” the proposal.

2025-10-02 13:02:37 UTC

- reply

A final note: it’s not totally clear what the UK is asking for here, but one theory is: they want access to foreign users who have ADP encryption on, but are on UK soil (for example, foreign diplomats.)
This would be a technical nightmare, but it fits the facts.
Would be a technical nightmare because now you’d have two versions of Apple encryption, where the security changes depending on where in the world you are (ie your IP address.) This is so dangerous that it’s tantamount to a global backdoor, including against US citizens.

2025-10-02 13:01:56 UTC

- reply

As to question (1), the article isn’t super clear. But it’s worth pointing out that Apple doesn’t just provide end-to-end encryption for backups through their ADP feature. They also provide end-to-end encrypted backup for health data, web history and passwords, even without ADP.
So “we want a backdoor to access iCloud encryption” could mean access to those password vaults and secure health data. This means Apple would have to disable those features in the UK as well, or deploy new insecure encryption code that could affect users globally.
It’s also possible that the UK is asking for a backdoor in Apple’s iMessage text messaging system. That isn’t technically “iCloud” but it’s adjacent enough that I could see it getting presented that way. This would not be the first time the UK gov expressed interest in that.
What’s worrying here is that the UK government seems absolutely determined to access user private data, no matter the bad press and the consequences. And they’re now willing to do it overtly. Between this and recent moves against encryption in the EU, we’re going to a bad place.

2025-10-02 13:00:38 UTC

- reply

So here we are again. This time the UK is narrowly targeting their request at UK users only. But this is baffling for two reasons.
First, Apple no longer offers the ADP encrypted backup feature to new UK users, and existing users are being migrated. So what is the UK asking for?
Second, how is Apple supposed to make a single encryption system that’s secure for US users but not for UK users? This would require that they partition their software globally and have many versions, which seems like a disaster. Are you sure you’re running the “secure” OS?

2025-10-02 12:59:57 UTC

- reply

For context: last year the UK demanded (in secret) that Apple create a backdoor into their iCloud Advanced Data Protection system. This was done via a Technical Capability Notice, and could have given the UK govt a master key to all iCloud users’ data worldwide.
When this fact became public, it drew a huge backlash. Apple announced that it was entirely removing the encrypted backup feature for all UK users. Trump administration figures rebuked the UK government.

2025-10-02 12:53:14 UTC

From yesterday: the UK is demanding another backdoor in Apple’s encryption. https://www.ft.com/content/d101fd62-14f9-4f51-beff-ea41e8794265?accessToken=zwAGQBl4pP6YkdPRAf1iFPlPUdO-_-pB6HlCZQ.MEYCIQDqJlwcbpQ4rrKlgeSJtbcTgqpW4uTX3yGMCHf2gdS0fQIhAMtU15LHqeHwAXTZ3wWJLDzI72kjsauSKc8OEDpim1Gx&sharetype=gift&token=3767767d-fe68-4911-a97e-c067715e061e

2025-09-27 13:38:43 UTC

“Industrial-scale pig butchering scams” god please let’s just skip this century and move on to the next.

2025-07-06 12:50:18 UTC

This battle will keep playing out over and over again until they achieve something that their own citizens have made it clear they don’t want. https://www.techradar.com/vpn/vpn-privacy-security/the-eu-wants-to-decrypt-your-private-data-by-2030

2025-04-02 19:06:01 UTC

A good story in WIRED giving updates on the situation with Dr. Xiaofeng Wang. https://www.wired.com/story/xiaofeng-wang-indiana-university-research-probe-china/

2025-03-31 16:25:05 UTC

The Indiana University chapter of the American Association of University Professors have written a letter demanding Dr. Xiaofeng Wang’s reinstatement.
https://files.ioc.exchange/media_attachments/files/114/257/972/308/019/391/original/551d06df173a4797.png

2025-03-30 01:05:14 UTC

I am jumping down with frustration at our academic community. People: we cannot do anything if everyone is unaware of professors getting arrested for multiple weeks.

2025-03-30 01:02:56 UTC

- reply

Some early non-Indiana coverage of this story. https://talkingpointsmemo.com/edblog/sketchy-first-reports

2025-03-29 20:14:22 UTC

- reply

Has anyone been in contact? I hear he’s been missing for two weeks and his students can’t reach him. How does this not get noticed for two weeks???

2025-03-29 20:11:42 UTC

- reply

Update to this story: I’ve now heard reports that Wang’s students have been unable to contact him since mid-March. His website was taken down on the 19th.

2025-03-29 20:09:44 UTC

- reply

this isn’t normal Karl.

2025-03-29 17:41:28 UTC

- reply

None of this is in any way normal.

2025-03-29 14:22:36 UTC

- reply

Xiaofeng’s profile is no longer available on IU sites. So here’s his Google Scholar. https://scholar.google.com/citations?user=pONu-5EAAAAJ&hl=en
https://files.ioc.exchange/media_attachments/files/114/246/166/266/585/875/original/0696ec39a4b0d9fa.jpeg

2025-03-29 14:20:48 UTC

- reply

House belonged to Xiaofeng Wang.
https://files.ioc.exchange/media_attachments/files/114/246/158/848/422/502/original/7062e8df68a62aca.jpeg
https://files.ioc.exchange/media_attachments/files/114/246/159/194/957/742/original/104f6855366351e6.jpeg

2025-03-29 14:14:57 UTC

I just heard that a cryptography professor at Indiana University had his house raided and was fired. Don’t know much more. https://www.heraldtimesonline.com/story/news/local/2025/03/28/fbi-department-of-homeland-security-agents-search-house-in-bloomington-indiana/82710451007/

2025-03-25 09:34:32 UTC

- reply

So there is one Signal Foundation doing the work that a dozen companies should be doing. No idea what to do about any of that.

2025-03-25 09:34:06 UTC

- reply

I guess my takeaway (1) is: no matter what people say, actual privacy is one of the most valuable services in the entire world, (2) network effects ensure a winner, yet (3) it is a totally unstable balancing act for for-profit companies to provide this, long term.

2025-03-25 09:33:17 UTC

- reply

I have too many reasons to worry about this but that’s not really the point. The thing I’m worried about is that, as the only encrypted messenger people seem to *really* trust, Signal is going to end up being a target for too many people.
Signal was designed to be a consumer-grade messaging app. It’s really, really good for that purpose. And obviously “excellent consumer grade” has a lot of intersection with military-grade cryptography just because that’s how the world works. But it is being asked to do a lot!
Right now a single technical organization is being asked to defend (at least) one side in a major regional war, the political communications of the entire US administration, the comms of anyone opposed to them globally, big piles of NGOs, and millions of “ordinary” folks to boot.
(There is no such thing as “ordinary user” cryptography BTW. Those ordinary users include CEOs, military folks, people doing many-million-dollar crypto trades through the app, etc. It’s a lot to put on one app and one non-profit.)
On top of this, it’s only a matter of time until governments (maybe in the US or Europe) start putting pressure on the infrastructure that Signal uses — which is mostly operated by US companies. I’m not sure how this will go down but it’s inevitable.

2025-03-25 09:31:38 UTC

You should use Signal. Seriously. There are other encrypted messaging apps out there, but I don’t have as much faith in their longevity. In particular I have major concerns about the sustainability of for-profit apps in our new “AI” world.

2025-03-13 16:56:58 UTC

- reply

This maybe confirms a bad feeling I was getting a few weeks back. https://ioc.exchange/@matthew_d_green/113929669316823012

2025-03-13 16:50:02 UTC

- reply

Money quote(s) from the article. Note that there is no “you can’t deny” clause in the U.K. law.
https://files.ioc.exchange/media_attachments/files/114/156/148/995/630/237/original/5153343597264996.jpeg

2025-03-13 16:49:18 UTC

- reply

Senator Wyden asked both companies to confirm or deny that they’d received TCNs. Both parties said, essentially, “if we got one we wouldn’t be allowed to tell you under U.K. law.” https://www.wyden.senate.gov/news/press-releases/bipartisan-members-of-congress-to-uk-spy-court-uk-gag-orders-for-surveillance-backdoors-threaten-americans-security-and-privacy-impede-congressional-oversight
https://files.ioc.exchange/media_attachments/files/114/156/146/194/088/062/original/a69c15213c21a1f6.mp4

2025-03-13 16:48:08 UTC

So I am getting the distinct feeling that Google (in addition to Apple) got a Technical Capability Notice from the U.K.

2025-03-12 12:08:46 UTC

- reply

I want to add a re-post of a blog post I wrote about the risks of AI combined with end-to-end encryption systems. https://blog.cryptographyengineering.com/2025/01/17/lets-talk-about-ai-and-end-to-end-encryption/

2025-03-11 12:18:10 UTC

Meredith Whittaker on the security and privacy risks of Ai agents in messaging platforms, which I also worry about a lot. https://techcrunch.com/2025/03/07/signal-president-meredith-whittaker-calls-out-agentic-ai-as-having-profound-security-and-privacy-issues/

2025-03-01 17:03:52 UTC

So here’s a simple request to Apple. Apple iMessage needs to enable “disappearing messages.” And they need to do it soon. https://blog.cryptographyengineering.com/2025/03/01/dear-apple-add-disappearing-messages-to-imessage-right-now/

2025-02-23 16:51:40 UTC

Three questions about Apple, Encryption, and the U.K. https://blog.cryptographyengineering.com/2025/02/23/three-questions-about-apple-encryption-and-the-u-k/

2025-02-21 20:40:25 UTC

- reply

And the fact that there’s more at stake here may be important to you, since *your data* might be what’s at stake. //

2025-02-21 20:39:53 UTC

- reply

If you’re Apple in this situation — keeping in mind that you’re dealing with laws that mandate total secrecy, and a government that mixes national security and criminal law — what’s your first move? My guess is that your best strategy is to flip the table.
After all, if you disable the encryption feature at issue from all U.K. customers, to some extent the issue appears moot. (It’s not moot, of course. The U.K. would still be asking for access to non-U.K. users.) But it gives you a place you can work from.
So I don’t know that this is where Apple is coming from. I *do* know that, despite some deserved criticisms (many from me) Apple has never seemed like a company that just wants to submit and turn off encryption. So I want to propose that maybe, just maybe there’s more here.

2025-02-21 20:38:57 UTC

- reply

Let’s imagine the following hypothetical. Apple is not asked by the U.K. government to compromise encryption features for only U.K. users. They are asked, by the U.K. government, to compromise encryption features for iCloud users in countries all over the world, including the US.
This would be an absolutely crazy request, by the way. It would put the U.K. in the position of controlling the maximum level of security available to any user anywhere in the world, and doing it *secretly*. If you engage with that request, there is no coming back.

2025-02-21 20:38:19 UTC

One thing I’m seeing re: today’s Apple news is a lot of people blaming Apple. Saying that Apple is selling people out by withdrawing encryption features from the U.K. market. For example, here’s Tim Sweeney. I want to propose a different take.
https://files.ioc.exchange/media_attachments/files/114/043/800/509/313/265/original/e5e5f79d0da3c341.jpeg

2025-02-21 15:43:53 UTC

- reply

This move will not affect:
iMessage encryption (though backed-up messages will now be available at Apple)
iCloud Keychain
FaceTime
Health data
Other services like iCloud Backup and Photos and text message backups will not be end-to-end encrypted.

2025-02-21 15:39:53 UTC

- reply

Additionally:
"Apple can no longer offer Advanced Data Protection (ADP) in the United Kingdom to new users and current UK users will eventually need to disable this security feature. ADP protects iCloud data with end-to-end encryption, which means the data can only be decrypted by the user who owns it, and only on their trusted devices. We are gravely disappointed that the protections provided by ADP will not be available to our customers in the UK given the continuing rise of data breaches and other threats to customer privacy. Enhancing the security of cloud storage with end-to-end encryption is more urgent than ever before. Apple remains committed to offering our users the highest level of security for their personal data and are hopeful that we will be able to do so in the future in the United Kingdom. As we have said many times before, we have never built a backdoor or master key to any of our products or services and we never will.”

2025-02-21 15:23:47 UTC

Apple has yanked Advanced Data Protection in the U.K. https://www.bbc.com/news/articles/cgj54eq4vejo

2025-02-15 13:27:28 UTC

Seems like a lot of the NIST standards have gone missing.

2025-02-13 13:23:40 UTC

Legislators in the US are pushing back on the UK’s demands for backdoors in Apple encryption. https://archive.is/2025.02.13-120308/https://www.washingtonpost.com/technology/2025/02/13/apple-uk-security-back-door-adp/

2025-02-12 19:09:29 UTC

I wrote a bit more about the UK’s recent move to allegedly demand backdoors in Apple encryption. https://blog.cryptographyengineering.com/2025/02/12/u-k-asks-to-backdoor-icloud-backup-encryption/

2025-02-11 02:40:15 UTC

Following up on the recent news about the U.K. government attempting to force Apple to backdoor its cloud encryption, Alex Stamos and I wrote this op-ed in the WSJ. https://www.wsj.com/opinion/u-k-kicks-apples-door-open-for-china-encryption-data-protection-deb4bc2b?st=uG1z3q&reflink=article_copyURL_share

2025-02-07 12:23:50 UTC

- reply

Apple hinted at this in a filing it made over a year ago to the U.K. government. https://publications.parliament.uk/pa/cm5804/cmpublic/InvestigatoryPowersAmendment/memo/IPAB10.htm
https://files.ioc.exchange/media_attachments/files/113/962/583/704/564/535/original/4ec78e6bd31371c3.jpeg

2025-02-07 12:23:05 UTC

- reply

The insane thing about the U.K. law is that it does not only apply to U.K. customers. It can potentially be used to go after non-UK customers as well. Say, people here in the US.

2025-02-07 12:22:31 UTC

- reply

Let’s be clear about what this article is saying. The U.K. has a law that allows it to issue “technical capability notices” to companies. These notices require the company to effectively disable, or secretly backdoor, their encryption mechanisms.

2025-02-07 12:21:37 UTC

The U.K. may be preparing to issue Apple an order that forces them to (secretly) disable encryption. https://www.washingtonpost.com/technology/2025/02/07/apple-encryption-backdoor-uk/

2025-02-01 17:00:15 UTC

- reply

I wish I could offer more than a funny feeling. I think the upshot of all of this is that the companies doing “opt in” backup should be pushing harder to make “opt in” into “heavily encouraged” and then into “by default.” The sooner the better.

2025-02-01 16:58:22 UTC

- reply

So maybe it would be good to give a lay of the land on this issue. Here is what I know about fully-E2E backup in major services:
Apple iCloud: available as an opt-in (called ADP)
Google: on for Android backups, if you use Google/Android backup (caveats)
Meta/WhatsApp: available opt-in, sometimes by default (for texts)

2025-02-01 16:53:47 UTC

- reply

This is a company that has cloud backup E2E encrypted as a default. And so at first I assumed that this was just a business demand — that it was costing the company a lot to keep this service running. But now I’m getting more worried about it. Maybe I’m paranoid.

2025-02-01 16:53:12 UTC

- reply

So why am I tweeting about this on a Saturday? Because something funny happened recently.
I heard (thirdhand) from a person at Big Company X that they were under pressure to disable end-to-end encryption for cloud backup.

2025-02-01 16:38:46 UTC

- reply

The one place where end-to-end encryption is “weakest”, ie where deployment rates are lowest, is *cloud backup*. This is, coincidentally, one of the best places for governments to obtain data.

2025-02-01 16:36:53 UTC

- reply

Local device encryption seems to be ubiquitous now. It’s really hard to see a country demanding that phones become unencrypted. There would be huge pushback globally. So again, where is the weak link that governments can push on?

2025-02-01 16:36:37 UTC

- reply

The question is *what*. Encrypted messaging seems to be proliferating and there are now many alternatives, from Apple, Meta, Signal. Many of those companies have threatened to leave countries that ban it. I have a hard time seeing any US-aligned country wipe that out.

2025-02-01 16:36:19 UTC

Things have been relatively quiet on the “crypto wars” front, which makes me think we’re going to see something dramatic soon. Maybe not here in the US, but probably from another US-allied country.

2025-01-17 15:43:55 UTC

I wrote a post about how AI will interface with end-to-end encryption. TL;DR maybe not so well! https://blog.cryptographyengineering.com/2025/01/17/lets-talk-about-ai-and-end-to-end-encryption/