I don't know everything there is to know about passkeys or anything, but my reaction to most of these comments is "You're passing yourself off as someone who has relevant opinions about security, and you can't possibly imagine how these things work or how they're useful? Come the fuck on."

Passkeys are basically session cookies that are signed by a secure element in one of your devices at the time you log in. That's it. They cannot be phished because there is no password to steal. If I had to guess, the basic flow is something like this:

1. When the passkey is created, the device's secure element coughs up a public key or something to the server representing itself as a trusted device

2. When a user logs in, the server issues a challenge (basically a random number) to the device and says "sign this with a private key that corresponds to one of the trusted public keys I have"

3. The secure element signs the challenge and sends it back

4. The server goes through its list of trusted device public keys until it finds one that verifies the challenge response. If it finds one, it logs you in. If it doesn't, you don't log in

Step (1) is probably bootstrapped with a username/password and second factor like SMS 2FA, OTP, or email 2FA.

Even if this isn't exactly how they work, it's a plausible implementation. Nothing about this requires vendor lock-in. The various secure elements that can produce passkeys come from many different places, so I'm sure sufficiently motivated open source people could create a firmware TPM that is certified for use with passkeys or something if they cared enough.