I don’t think the proposed PR is technically correct. The quantum-vulnerable 
algorithms secp256r1, secp384r1, X25519, and X448 are also marked 
RECOMMENDED=Y. Moreover, the fact that TLS currently recommends standalone 
quantum-vulnerable key exchange more strongly than standalone quantum-resistant 
key exchange is probably not something we want to codify in an RFC. The purpose 
of the IANA RECOMMENDED column is precisely to avoid embedding such these kind 
of policy decisions in RFC text.

I find the existing text in the draft quite good, except that "PQ/T" will not 
age well. X25519MLKEM768 is already the de facto standard, so I do not think it 
needs much additional promotion. My takeaway from the recent discussions is 
that there is a significant risk that people will assume PQ/T provides more 
security and robustness than it actually does. While PQ/T can continue to be 
used for many years, we are quickly approaching the point (2029?) where the 
IETF should stop implying that the traditional component contributes meaningful 
security.

Cheers,
John Preuß Mattsson

From: Filippo Valsorda <[email protected]>
Date: Monday, 8 June 2026 at 01:27
To: [email protected] <[email protected]>
Subject: [TLS] Re: FATT Chance: On the Robustness of Standalone and Hybrid 
ML-KEM Key Exchange in TLS 1.3

2026-06-08 01:05 GMT+02:00 Muhammad Usama Sardar 
<[email protected]<mailto:[email protected]>>:

[0] https://github.com/tlswg/draft-ietf-tls-mlkem/pull/21

I agree with the point made by others before that a spec for X doesn’t have to 
compare it with Y, or reproduce the rest of the IANA registry Recommended 
column, so I disagree with this PR.
_______________________________________________
TLS mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to