With a password manager storing unique passwords, your password is still being sent over the network to that website each time you log in.
With a password manager storing passkeys, your private passkey is not transmitted as part of the website request. It can't be intercepted or accidentally/negligently stored in plaintext in a database or server logs.
You still have to trust the secure syncing of your passkeys, just like you do with passwords, but there are still fewer threat vectors than with passwords.
It’s more than you think: if I can convince you to type in your Google password at g00gl3.com I can turn around and use that password for the real server. Passkeys block that attack.
So they couldn't convince everyone to give up their privacy and use Facebook or Google to track our logins to other websites via "Login With", and this is a new way?
If by new you mean a completely different mechanism which does not share the privacy considerations of a single sign on service, yes. If you care about privacy, passkeys are a much better option.
1. Passkeys are unique, so no more password stuffing attacks.
2. Passkeys can't be to short, in contrast to passwords
Passkeys essentially remove almost all risks for websites and moves them to the user (lost passkey, attacks on their password managers). It is not perfect, but it removes a lot of problems that we have right now (like more than a billion leaked passwords in the wild)
So it adds no value over the long, randomly generated passwords from a password manager besides making a new standard and giving Microsoft a reason to push a new dark pattern on users to force higher uptake rates of this new, bespoke standard with limited support?
When you log into a site with a password, you're needing to transmit your actual password over the wire. Sure, you'll hopefully have encryption when sending that credential over, but in the end you'll be exchanging your actual password with some remote service. That remote service might be misconfigured, it might be misdesigned, it might be under attacker control. Now they have your whole credential.
A passkey doesn't transmit your actual, full, repeatable credential over the wire. It's a challenge-response protocol, so only that one authenticated session would be intercepted. Kill all questionable sessions and you're good, they're not reusing it.
> randomly generated passwords from a password manager
are impossible to enforce. If you present users with password field, a sizable percentage of them will just manually type in the same weak, compromised password that they've used on every other site they've ever created an account on in their life. Passkeys are much harder to misuse. That's where 99% of their value is.
Yes there are also other advantages, like the fact that passkeys use public key cryptography, but those are tiny compared to the human factors improvements.
Passkeys include a key that the website you're logging in to holds, if a site can't present that key then the passkey doesn't work, meaning phishing attacks no longer work because 0utlook.com doesn't have the key that outlook.com holds.
It's a standard supported by multiple parties, not just Microsoft, including multiple open source password managers.
And it does provide some benefits: phishing protection (no shared secret that can be intercepted or given to the wrong party) and the service does not need to store as much sensitive information (don't need password hashes that could be leaked and cracked, just a public key).
Device attestation to allow blocking "undesirable" devices from authenticating and lock in purposes.
*keypass was working on an export feature and there were already threats to use the attestation club to ban them from the landscape for not falling in line
I found this out in October when trying to figure out this complaint.
timcappalli from FIDO Alliance mentioned in that above thread that plain text exports shouldn't be allowed, and that password managers/providers should be blocked if they implement plain text export.
Since that thread, there's a new spec that allows users to securely migrate passkeys from one provider to another, but no way to export to plain text (for debug purposes, or if there's a bug in the export/import and you need to troubleshoot, etc).
For me, threatening to block providers for implementing a feature that I desire is a great way for me to lose all interest in passkeys completely. I don't trust FIDO Alliance to make the right call nor do I trust big tech companies to produce bug-free software.
This very much falls into the same box as “not your keys, not your crypto”: if you’re forced to trust someone else to manage the keys for you then they have them - necessarily, in order to permit “transfer” (under this scheme, not everything) to another party - in plaintext, while you’re not allowed to “for your own good”, then you’ve lost it all.
They can:
1. Impersonate you, gaining access to anything your keys unlock
1.a. Impersonate you, claiming to be you in a violation of “key use enables non-repudiation”
2. Deny you the ability to use your keys
2.a. Change any of your keys, locking you out of things
3. Deny you the ability to transfer your keys to anyone they “don’t like”
3. Provide your keys to anyone else, e.g. “with a court order”
3.a. Anyone “benefitting” under (3) can then do (1(a))
…and surely more Bad Things.
Every single time “passkeys” seems to like “okay, maybe”… some fucktards pull another one of these.
Then I go, “okay, ssh keys, PIV, or whatever else is Just Fine, and these people who are either state agents, idiots, or power hungry idiots working to advance total control over humans with lack of freedom and no way back can go die, or as an alternative be sentenced to serious computer-things-reeducation”. …and I kinda mean it. There are certain things you just don’t come back from, as a society, etc. and I just won’t support it. You only get one chance not to.
Passwords are also a standard supported by pretty much everyone, and password managers (including those built into browsers) already generate long, unique, phishing resistant (it only prompts on the matching domain) passwords.
The main difference I see with passkeys from a usability standpoint is that Firefox doesn't have built-in support for a software implementation, making them literally unusable for me.
Firefox does support passkeys but their native implementation is behind a feature flag. Beyond that Firefox add-ons (such as those for password managers) can enable support for their own purposes.
For example 1Password can be used for passkeys in Firefox.
I can't find that option's documentation. Do you have a link? The only documentation I've seen indicates that they only support hardware devices, and I don't own one.
~90% of people who are presented with a password input field are not and never will be "password manager users". 100% of people presented with a passkey prompt are, because the very nature of passkeys is that they're stored in what is effectively a password manager.
Ah, I think I misunderstood that comment. From a global perspective "the benefit" compared to using passwords in a password manager is indeed exactly what I said, but I guess that user was asking more about the personal benefit on an individual level.
From that perspective it just makes the UX slightly smoother and makes it impossible for the site to screw up and leak your plaintext creds. Other than that yeah there's not a big difference compared to using an autofilled, unique, randomly generated password. Which is good, because eventually sites are going to start phasing out that latter option for the exact reasons I outlined in my previous comment.
>> Passkeys are unique, so no more password stuffing attacks.
> Just like passwords. What is the difference ?
>> Passkeys can't be to short, in contrast to passwords
> So a "long password" is a "passkey" ?
Of course not.
Passkeys are effectively just key pairs defined by a FIDO standard. It’s much more productive to think of passkeys as mutual certificate authentication designed for use by the masses.
If you’ve ever used a Yubikey for primary authentication, you’ve already used a passkey.
I really don't know how you're able to come up with the idea, that passwords are unique. They are in no aspect whatsoever unique. You as a user can try to use passwords, that are unique to you. That doesn't (can't and shouldn't) mean, that those passwords are unique against the database of a website, or other services.
If the password manager is unlocked, I get logged in automatically anyways.