PQC / post quantum version of HAIP

[jogu]

At some point we'll likely need to publish a post-quantum safe version of HAIP, which I think mostly means mandating whichever quantum safe JOSE algorithm seems best for signing requests / encrypting responses. Don't think it needs any action right now, just creating the issue so there's a place to record any thoughts people have.

[c2bo]

It would be extremely useful to understand the stance of different regions on this (pure PQ, hybrid signatures and which schemes) and especially planned hardware support for smartphones (SEs) for the device binding part of it. Since it might take a bit of time to collect inputs, it's probably a good idea to start this early?

[fkj]

I agree this is something we need to do (as in all other OpenID specs that have default crypto values).

However, I think we need to consider what we want HAIP to be. Should HAIP be a best practice or should it aim to support all different regulatory regimes? This will make a big difference in particular regarding whether we support hybrid schemes.

[jogu]

Martijn mentioned on today's call that we also need to consider the credential formats here - SD-JWT VC/mdoc profiles potentially need updates for PQC algorithms for signing the credential and the device signing.

Discussion with Dima/Frederik/Martijn about this being 1.2 or later.