Support was first offered in May, but now Google suggests it as the default choice.

Passkey is some sort of specific unique key to a device allowing to use a pin on a device instead of the password. But which won’t work on another device.

Now I don’t know if that key can be stolen or not, or if it’s really more secure or not, as people have really unsecure pins.

  • @DeadlineX@lemm.eeEnglish
    7•1 year ago

    Mostly phishing. Passkeys can’t be phished. And really, passwords are awful in general for security purposes. You don’t have to use your phone or google or apple or whatever.

    I actually have a physical usb key that I use as a passkey. Its just a more secure login implementation and will likely be the only option in the future.

    • @hedgehog@ttrpg.networkEnglish
      2•1 year ago

      Passkeys can be phished, it’s just much more difficult than with passwords, TOTP MFA, SMS MFA, other OTPs, or push notification-based MFA (e.g., Duo or the way Microsoft, Apple, and Google push a notification to their app and you confirm and/or enter the key).

      Passkey is extremely phishing resistant in the same as Webauthn MFA and U2F MFA are, in that origin checks by the browser prevent attackers from initiating the auth process. But it can still be attacked in these ways:

      1. XSS bug in the target website
      2. Browser vulnerability
      3. Malicious browser (not a concern on iOS but a concern everywhere else)
      4. Compromise of any cert in the chain between you and the target website
      5. Convincing the user to install (or using malware to install) a root certificate, or compromising one you already installed (e.g., for work)
      6. Bookmarklet/clipboard/devtools attacks

      From memory, passkeys, webauthn, and u2f should prevent over 99% of phishing attacks that are successful without them in place.

      There’s also the risk of the passkey itself being compromised, though that level of risk is dependent on your device / how you’re storing your passkeys and isn’t a “phishing” risk.

      • NatanaelEnglish
        2•1 year ago

        The main point is all those attacks need to attack the local software or hardware implementation on one of the two ends (or a cert issuer), and even then it’s replay protected so for example an XSS attack lasts only for one session, so it’s more robust.