The very last thing, past the conclusion, the very last footnote is:
> Addendum: I feel I have made a compelling case for passkeys in this article. And I haven’t even pointed out one of their best security features: they are (in most use cases) domain-bound, meaning they are resistant to what are called Man-In-The-Middle attacks.
"I made a compelling case... but didn't point out the most important security feature that motivated passkeys in the first place"
Honestly I don't think this post did make a compelling case. There's a lot of assuming bad faith on the part of the FIDO alliance, while completely missing that preventing MITM attacks was a key design feature. The FIDO alliance members deal with stolen accounts all the time, and passkeys largely solve that. Stealing accounts is almost always done with phishing or social engineering, and passkeys solve the former and reduce the surface for the latter.
I don’t know if it’s productive for me to talk about how a headline is off-putting, but I’m going to mention it because, well, it’s a headline, and a headline is a strong first impression.
Is the audience of this article the kind of people who are already using passkeys and like them? Because, if so, we’re starting at a deficit here. The author is out to get my beloved passkeys that have saved me so much time and pain.
Are passkeys perfect? No. Are they even very good for no-technical users? Absolutely not…they’re incredibly confusing for that kind of person if you ask me.
However, for my workflow, I’ll take them over a password+2FA type of situation where logging in is such an annoyance. The security part of the discussion is very interesting and a lot of the things in this article are stuff that I didn’t know before reading it.
The issue is that, putting on my user hat, I just don’t care. As long as my accounts are reasonably secure I’m much more concerned about ease of use and workflow.
The intended audience is people who are not yet using passkeys, or have just started (due to all the websites demanding them now) but don't really know what they are.
I apologize that the headline is provocative. I thought for a long time about how to appeal to both people who are barely aware of what passkeys are but also not put off people who like them. I considered calling them a "Trojan Horse" in the headline as I do in the body (i.e. legitimately lovely, but with a terrible and unobvious problem) but "trojan" has another connotation in computer security.
Nonetheless, I said what I meant. Convenience is how they get you. This convenience didn't have to come with a trap, but thanks to the way it was designed, it does.
Sure, I think your logic makes sense, and I don’t want to overemphasize complaints about headlines.
Regarding how I feel about passkeys after reading your article, logically I am reading what you wrote and conceptually understand the trap, it’s just hard to wrap my head around what kind of situation would cause this trap to “catch” me so to speak.
I.e., the upside is real and immediate, the downside is theoretical and tied to future uncertainty.
This is a (very long and) well-written treatise against passkeys in their common and disparate forms. It's worth reading, and brings up several interesting and often concerning points.
I don't agree with the conclusion, though.
TOTP is great, flexible, and keeps the user in control; however, using a passkey in a flexible password manager (Bitwarden, keepass, 1Password, etc) really does continue to give the user control. And the user experience is dramatically easier than filling in TOTP codes.
To advocate for TOTP over passkeys, we're already relying on an application to generate those numbers (authenticator apps, password managers, etc). I see no reason to not simply continue using those pieces of software to manage passkeys, too.
I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys. Passwords are simply unparalleled in their flexibility. You're never going to be locked out because <bluetooth, your TPM, your phone, ...> isn't working today.
I think the bit about TOTP was mostly a joke, the problem with passkeys is that they're billed as a replacement for passwords but they mix lots of MFA concerns in and make interoperability essentially impossible a lot of the time.
> I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys.
That's simply not true. I'm able to log onto GitHub on my corporate device using a passkey that's kept in 1Password on my iPhone.
But you can be locked out if the provider decides you need to change your password after a breach, or you forget the password, or you enter the wrong password in too many times.
They’re both equally disposable. You can reset a password just like you can reset a passkey.
I completely agree that KeePassXC and other password managers give you total user control, and they do support "passkeys" -- but crucially, not all kinds of passkeys. KeePassXC has no remote attestation support, nor hardware-backing support, and nor do I think it should.
My claim is not that passkeys couldn't give you control in principle. It's that the standard readily allows for websites to prevent you from being allowed to exercise that control, and the unclear messaging and UX around passkeys means most users will not even notice should this happen.
Note that attestation can only be required during credential registration; at login time it's not possible to retroactively require attestation if it wasn't obtained during registration. And browsers will throw up a permission prompt if a website requests attestation. Are Netflix and Twitter currently requesting attestation when people register passkeys, and rejecting the registration if the user declines the permission prompt? I very much doubt it, since Apple's passkey implementation doesn't support attestation.
Therefore, the author's scenario that sites will wait until most of their users are using attested passkeys and then "flip the switch and start rejecting non-attested passkeys" isn't very plausible.
Attestation is certainly a stain on the WebAuthn spec, and it would have been better if Chrome and Firefox didn't support attestation (unless perhaps the browser is enrolled in enterprise management), but this blog post is presenting a hypothetical and unlikely scenario. That's not a good enough reason to recommend TOTP instead, given that TOTP is vulnerable to phishing and passkeys aren't.
If consumer-facing sites start requesting attestation when registering passkeys, they deserve to be vigorously called out, and maybe then the recommendation to use passkeys should be revisited. But until that happens, don't tell people not to use passkeys.
I've yet to voluntarily enable passkeys, for the main reason that I want portability, and not to be tied to any single device that can be lost. I want to be able to backup and lock my credentials in a safe. I also want to control the strength of my credentials for different services, and I don't consider my thumbprint used to unlock my phone to be as strong as a long password in a password manager.
The first time I was forced to use a passkey implemented within the vendor's app (3rd party password managers were not an option). I quickly closed that account.
The second time was an app that popped up a passkey opt-in in the middle of a bunch of transaction screens. I quickly realized the error, but there was no easy undo. The opt in was one green button in the middle of the screen. Turning it back off required digging through settings to find the security option to disable, and then confirm my choice multiple times. That process also signed out all my other devices.
The fact that companies are resorting to dark patterns and forced requirements, where my money is held hostage, should be pretty good evidence of how poorly the passkey rollout is going.
You can put your passkey in another safe. I store mine in my vaultwarden. You can also do keepass with a plugin. You don't have to store it on your phone.
Ideally, passkeys should be cattle, not pets. You should have more than one per account, saved to more than one password manager. Then it's fine if they're not copyable because you can get another.
I'm sorry, what? Who has more than one password manager (other than a personal/work split)? That's absurd and not something anyone I've heard of would put up with
I've been complaining about passkeys for a long time and it was driving me crazy that it seemed to be the one thing that Big Tech, tech-savvy people, and normies were all on the same page about. The recent PR push a few months back was overwhelming.
Passkeys in theory are great, but the capabilities of the spec are worrying. Defaults can and do change, and when they do, so follows 99% of the population, and then you find yourself either getting blocked out of services because your hardware/software doesn't comply, or you give in.
But on the other end, it might not matter anyways. Apple, Google, Cloudflare, are already pushing similar garbage and now you increasingly have to scan a QR code that verifies your mobile hardware to use a desktop.
I'm in small tech saas and users having their accounts breached because they share a password with 100 other services is a constant. I have had to set forced 2FA on many products because we take the blame when the user has their account compromised.
Passkeys offer the same and more security but are less of a pain for the user.
If you also support TOTP and physical security keys as an alternative to passkeys, that should be enough to provide a stable long-term alternative for users of FOSS devices. And of course, RP-side you can choose to never require the remote attestation flag to be set.
Convenience is precisely the thing that big tech established tech platforms want to reward their users with to keep them around. The undeniable convenience of passkeys didn't have to come with a standard that is threatening against unattestable devices like my Linux phone. But it does, and that means passkeys are a way that the industry can discourage people away from using non-Google/Apple/Microsoft-approved devices.
Passkeys make it feel like the site is trying to get more info about me; it seems harder to maintain multiple accounts with passkeys. They feel worse for anonymity than a password where you know exactly what you're pasting into the site.
Although this isn't true because actually with a password you have to choose how to generate it and everyone chooses different ways.
My understanding is that TOTP and physical security keys both share very little information about you. Perhaps passkeys, being part of a standard that the FIDO Alliance keeps updating, may some day be expected to reveal considerably more information. And the limited number of remote-attestation-approved passkey managers might someday all start start volunteering such information regardless.
Because you or someone needs to publish "<digital_signature>" in a way that cannot be forged, stolen, MiTM'd, or otherwise compromised. Basically the key generating that signature just becomes another password, far more difficult to use in practice, and sharing all the same disadvantages.
So no one implements it for site-local login. Even the biggest sites like Amazon, for example, still use old-school passwords+2FA.
The way secure auth works for small sites is that a third party[1] authenticates you, who the website trusts more than mere users. And that's where all the complexity comes up. You need something secure stored on your known-secure device, a password alone won't do.
[1] In practice Google or Meta. Occasionally Apple or Microsoft. Everyone else is noise.
> What’s important to know is that unlike with any other kind of login credential, you do not have ultimate control over your passkeys.
Use proper password managers, like keepassxc. Then you have control. But problem is that not all sites support that. If you can't use the above - then yes, passkeys are bad, especially if they become mandatory.
If you keep reading the article, you’ll see that the claim is current (unattested) passkeys are a stepping stone to “attested passkeys”. Attested passkeys would require remote attestation with OS, TPM, and password manager all working together, and would allow websites to require things like hardware backed non transferable passkeys that you don’t control. This has always been my suspicion and I do think it’s worth worrying about.
The very last thing, past the conclusion, the very last footnote is:
> Addendum: I feel I have made a compelling case for passkeys in this article. And I haven’t even pointed out one of their best security features: they are (in most use cases) domain-bound, meaning they are resistant to what are called Man-In-The-Middle attacks.
"I made a compelling case... but didn't point out the most important security feature that motivated passkeys in the first place"
Honestly I don't think this post did make a compelling case. There's a lot of assuming bad faith on the part of the FIDO alliance, while completely missing that preventing MITM attacks was a key design feature. The FIDO alliance members deal with stolen accounts all the time, and passkeys largely solve that. Stealing accounts is almost always done with phishing or social engineering, and passkeys solve the former and reduce the surface for the latter.
I don’t know if it’s productive for me to talk about how a headline is off-putting, but I’m going to mention it because, well, it’s a headline, and a headline is a strong first impression.
Is the audience of this article the kind of people who are already using passkeys and like them? Because, if so, we’re starting at a deficit here. The author is out to get my beloved passkeys that have saved me so much time and pain.
Are passkeys perfect? No. Are they even very good for no-technical users? Absolutely not…they’re incredibly confusing for that kind of person if you ask me.
However, for my workflow, I’ll take them over a password+2FA type of situation where logging in is such an annoyance. The security part of the discussion is very interesting and a lot of the things in this article are stuff that I didn’t know before reading it.
The issue is that, putting on my user hat, I just don’t care. As long as my accounts are reasonably secure I’m much more concerned about ease of use and workflow.
The intended audience is people who are not yet using passkeys, or have just started (due to all the websites demanding them now) but don't really know what they are.
I apologize that the headline is provocative. I thought for a long time about how to appeal to both people who are barely aware of what passkeys are but also not put off people who like them. I considered calling them a "Trojan Horse" in the headline as I do in the body (i.e. legitimately lovely, but with a terrible and unobvious problem) but "trojan" has another connotation in computer security.
Nonetheless, I said what I meant. Convenience is how they get you. This convenience didn't have to come with a trap, but thanks to the way it was designed, it does.
Sure, I think your logic makes sense, and I don’t want to overemphasize complaints about headlines.
Regarding how I feel about passkeys after reading your article, logically I am reading what you wrote and conceptually understand the trap, it’s just hard to wrap my head around what kind of situation would cause this trap to “catch” me so to speak.
I.e., the upside is real and immediate, the downside is theoretical and tied to future uncertainty.
This is a (very long and) well-written treatise against passkeys in their common and disparate forms. It's worth reading, and brings up several interesting and often concerning points.
I don't agree with the conclusion, though.
TOTP is great, flexible, and keeps the user in control; however, using a passkey in a flexible password manager (Bitwarden, keepass, 1Password, etc) really does continue to give the user control. And the user experience is dramatically easier than filling in TOTP codes.
To advocate for TOTP over passkeys, we're already relying on an application to generate those numbers (authenticator apps, password managers, etc). I see no reason to not simply continue using those pieces of software to manage passkeys, too.
I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys. Passwords are simply unparalleled in their flexibility. You're never going to be locked out because <bluetooth, your TPM, your phone, ...> isn't working today.
I think the bit about TOTP was mostly a joke, the problem with passkeys is that they're billed as a replacement for passwords but they mix lots of MFA concerns in and make interoperability essentially impossible a lot of the time.
> I can't use a passkey on a device unless I can install my password manager on that device and I am comfortable giving that device access to my passkeys.
That's simply not true. I'm able to log onto GitHub on my corporate device using a passkey that's kept in 1Password on my iPhone.
https://bughunters.google.com/blog/passkeys
But you can be locked out if the provider decides you need to change your password after a breach, or you forget the password, or you enter the wrong password in too many times.
They’re both equally disposable. You can reset a password just like you can reset a passkey.
I completely agree that KeePassXC and other password managers give you total user control, and they do support "passkeys" -- but crucially, not all kinds of passkeys. KeePassXC has no remote attestation support, nor hardware-backing support, and nor do I think it should.
My claim is not that passkeys couldn't give you control in principle. It's that the standard readily allows for websites to prevent you from being allowed to exercise that control, and the unclear messaging and UX around passkeys means most users will not even notice should this happen.
Note that attestation can only be required during credential registration; at login time it's not possible to retroactively require attestation if it wasn't obtained during registration. And browsers will throw up a permission prompt if a website requests attestation. Are Netflix and Twitter currently requesting attestation when people register passkeys, and rejecting the registration if the user declines the permission prompt? I very much doubt it, since Apple's passkey implementation doesn't support attestation.
Therefore, the author's scenario that sites will wait until most of their users are using attested passkeys and then "flip the switch and start rejecting non-attested passkeys" isn't very plausible.
Attestation is certainly a stain on the WebAuthn spec, and it would have been better if Chrome and Firefox didn't support attestation (unless perhaps the browser is enrolled in enterprise management), but this blog post is presenting a hypothetical and unlikely scenario. That's not a good enough reason to recommend TOTP instead, given that TOTP is vulnerable to phishing and passkeys aren't.
If consumer-facing sites start requesting attestation when registering passkeys, they deserve to be vigorously called out, and maybe then the recommendation to use passkeys should be revisited. But until that happens, don't tell people not to use passkeys.
I've yet to voluntarily enable passkeys, for the main reason that I want portability, and not to be tied to any single device that can be lost. I want to be able to backup and lock my credentials in a safe. I also want to control the strength of my credentials for different services, and I don't consider my thumbprint used to unlock my phone to be as strong as a long password in a password manager.
The first time I was forced to use a passkey implemented within the vendor's app (3rd party password managers were not an option). I quickly closed that account.
The second time was an app that popped up a passkey opt-in in the middle of a bunch of transaction screens. I quickly realized the error, but there was no easy undo. The opt in was one green button in the middle of the screen. Turning it back off required digging through settings to find the security option to disable, and then confirm my choice multiple times. That process also signed out all my other devices.
The fact that companies are resorting to dark patterns and forced requirements, where my money is held hostage, should be pretty good evidence of how poorly the passkey rollout is going.
You can put your passkey in another safe. I store mine in my vaultwarden. You can also do keepass with a plugin. You don't have to store it on your phone.
Ideally, passkeys should be cattle, not pets. You should have more than one per account, saved to more than one password manager. Then it's fine if they're not copyable because you can get another.
> more than one password manager
I'm sorry, what? Who has more than one password manager (other than a personal/work split)? That's absurd and not something anyone I've heard of would put up with
Want to guess how many non-tech individuals save passwords in Chrome on desktop and in iOS on their iPhone?
At that point you don't have a password manager, you have a bunch of passwords you know and you just left "remember this" checked.
if I can't have two password managers for redundancy then none it is.
I've been complaining about passkeys for a long time and it was driving me crazy that it seemed to be the one thing that Big Tech, tech-savvy people, and normies were all on the same page about. The recent PR push a few months back was overwhelming.
Passkeys in theory are great, but the capabilities of the spec are worrying. Defaults can and do change, and when they do, so follows 99% of the population, and then you find yourself either getting blocked out of services because your hardware/software doesn't comply, or you give in.
But on the other end, it might not matter anyways. Apple, Google, Cloudflare, are already pushing similar garbage and now you increasingly have to scan a QR code that verifies your mobile hardware to use a desktop.
Knowing how big tech works this will get forced down throats desired or not
I'm in small tech saas and users having their accounts breached because they share a password with 100 other services is a constant. I have had to set forced 2FA on many products because we take the blame when the user has their account compromised.
Passkeys offer the same and more security but are less of a pain for the user.
If you also support TOTP and physical security keys as an alternative to passkeys, that should be enough to provide a stable long-term alternative for users of FOSS devices. And of course, RP-side you can choose to never require the remote attestation flag to be set.
Convenience is precisely the thing that big tech established tech platforms want to reward their users with to keep them around. The undeniable convenience of passkeys didn't have to come with a standard that is threatening against unattestable devices like my Linux phone. But it does, and that means passkeys are a way that the industry can discourage people away from using non-Google/Apple/Microsoft-approved devices.
Passkeys make it feel like the site is trying to get more info about me; it seems harder to maintain multiple accounts with passkeys. They feel worse for anonymity than a password where you know exactly what you're pasting into the site.
Although this isn't true because actually with a password you have to choose how to generate it and everyone chooses different ways.
My understanding is that TOTP and physical security keys both share very little information about you. Perhaps passkeys, being part of a standard that the FIDO Alliance keeps updating, may some day be expected to reveal considerably more information. And the limited number of remote-attestation-approved passkey managers might someday all start start volunteering such information regardless.
Can someone tell why the whole standard for authentication on the web is not just this:
me: Can you show me my stuff?
website: HTTP 401, who are you? I understand EasyAuth 1.0, if that is good for you.
me: Sure, here's my EasyAuth 1.0: Hi <website>, I'm <user_id>. The time is <timestamp>. Signed, <digital_signature>.
You have described, at a high level, passkeys.
Ok, I guess you also need a way to say "Oh, and next time I talk to you I'll use public key <public_key>."
Because you or someone needs to publish "<digital_signature>" in a way that cannot be forged, stolen, MiTM'd, or otherwise compromised. Basically the key generating that signature just becomes another password, far more difficult to use in practice, and sharing all the same disadvantages.
So no one implements it for site-local login. Even the biggest sites like Amazon, for example, still use old-school passwords+2FA.
The way secure auth works for small sites is that a third party[1] authenticates you, who the website trusts more than mere users. And that's where all the complexity comes up. You need something secure stored on your known-secure device, a password alone won't do.
[1] In practice Google or Meta. Occasionally Apple or Microsoft. Everyone else is noise.
I tried using a passkey to log in to something and it gave me an error and I simply didn't care about trying to fix it. So passwords it is for me.
> What’s important to know is that unlike with any other kind of login credential, you do not have ultimate control over your passkeys.
Use proper password managers, like keepassxc. Then you have control. But problem is that not all sites support that. If you can't use the above - then yes, passkeys are bad, especially if they become mandatory.
If you keep reading the article, you’ll see that the claim is current (unattested) passkeys are a stepping stone to “attested passkeys”. Attested passkeys would require remote attestation with OS, TPM, and password manager all working together, and would allow websites to require things like hardware backed non transferable passkeys that you don’t control. This has always been my suspicion and I do think it’s worth worrying about.