Skip to content
All news
News

IETF advances post-quantum MLS cipher suites

By Leon Acosta4 min read

Also mentioned

ML-KEMML-KEMModule-Lattice-based Key Encapsulation MechanismModule-Lattice-Based Key-Encapsulation Mechanism, the NIST-standardized post-quantum KEM derived from CRYSTALS-Kyber and specified in FIPS 203.Read the full entry (new tab), ML-DSAML-DSAModule-Lattice-based Digital Signature AlgorithmModule-Lattice-Based Digital Signature Algorithm, the NIST-standardized post-quantum signature scheme derived from CRYSTALS-Dilithium and specified in FIPS 204.Read the full entry (new tab), KEMKEMkey encapsulation mechanismA public-key mechanism for securely establishing a shared secret key: the sender encapsulates a random secret to the recipient's public key, and the recipient decapsulates it with their private key.Read the full entry (new tab), TLSTLSTransport Layer SecurityTransport Layer Security, the protocol that encrypts and authenticates most internet traffic, including HTTPS. It uses key exchange, certificates, and symmetric encryption to protect a session.Read the full entry (new tab), AES-GCMAES-GCMAdvanced Encryption Standard, Galois/Counter ModeThe Advanced Encryption Standard, a symmetric block cipher standardized by NIST that encrypts data using shared secret keys of 128, 192, or 256 bits.Read the full entry (new tab), FIPSFIPSFederal Information Processing StandardFederal Information Processing Standards, publicly announced standards developed by NIST for use in U.S. government computer systems, including cryptographic algorithms and modules.Read the full entry (new tab), NISTNISTNational Institute of Standards and TechnologyThe U.S. National Institute of Standards and Technology, the agency that develops and publishes cryptographic standards, including the FIPS series and post-quantum algorithms.Read the full entry (new tab), PQCPQCpost-quantum cryptographyCryptographic algorithms designed to run on today's classical computers while remaining secure against attacks by both classical and future quantum computers.Read the full entry (new tab) are defined in the glossary.

TL;DR

Read this first

The IETF has opened last call on eleven proposed cipher suites that add ML-KEM to Messaging Layer Security. Eight aim at post-quantum confidentiality while retaining traditional signatures. Three also use ML-DSA for post-quantum authentication. This is a standards milestone, not a released protocol update: draft-ietf-mls-pq-ciphersuites-08 can still change, has no assigned RFC number, and does not prove that any messaging product has deployed the suites.

On 19 November 2026, the IESG asked for community review of draft-ietf-mls-pq-ciphersuites-08 through 3 December. The document extends RFC 9420, the base Messaging Layer Security protocol for asynchronous end-to-end encrypted groups. Its practical change is a menu of ML-KEM based suites for the group key schedule, with both pure post-quantum and post-quantum plus traditional hybrid choices.

What changed in revision 08

Revision 08 entered IETF last call as a Proposed Standard candidate. The draft defines eleven suite registrations. They combine FIPS 203 ML-KEM with AES-GCM or ChaCha20-Poly1305, HKDF based on SHA-256 or SHA-384, and either traditional signatures or ML-DSA. Three hybrid KEM combinations pair ML-KEM-768 with X25519, ML-KEM-768 with P-256, or ML-KEM-1024 with P-384.

Suite groupKey establishmentAuthenticationClaim it can support
Eight confidentiality-focused suitesPure ML-KEM or a post-quantum plus traditional hybrid KEMTraditional signaturePost-quantum confidentiality against passive collection, not post-quantum identity authentication
Three full post-quantum suitesPure ML-KEM or ML-KEM-768 plus X25519ML-DSAPost-quantum confidentiality and authentication within the draft's stated assumptions
The proposed MLS choices separate key establishment from authentication.

That split matters. The draft states that the first eight suites protect confidentiality against passive quantum attackers but keep classical authentication. A future attacker able to forge the traditional signature or its credential could inject a KeyPackage or a group operation and compromise later epochs. Calling all eleven suites fully post-quantum would overstate the first eight.

Why MLS needs more than a new KEM identifier

MLS is not a two-party TLS handshake. A group maintains a ratchet tree, processes commits, derives epoch secrets and lets members update asynchronously. The cipher suite therefore affects HPKE, transcript hashes, the MLS key schedule, credential signatures and every client's ability to process the same group state. A migration must cover group creation, member addition, removal, update, external joins and recovery after a client misses epochs.

  • Inventory which MLS library and protocol version every client actually ships.
  • Confirm the exact suite identifier negotiated or selected for a new group.
  • Measure KeyPackage, Welcome and Commit sizes with realistic group membership.
  • Reject members whose credential or suite support does not meet the group's floor.
  • Test mixed-version groups, missed epochs, reinitialization and backup restoration.
  • Record whether confidentiality and authentication are each classical, hybrid or post-quantum.

The site's hybrid key exchange explainer now names MLS as a separate deployment surface. The same migration rule applies: document the combiner and downgrade floor rather than treating the presence of ML-KEM as proof of the final security property.

The draft depends on other work in progress

The last-call notice calls out normative downward references to four Internet-Drafts, including post-quantum HPKE, hybrid KEM construction and ML-DSA use in TLS. The MLS draft uses those definitions for HPKE KEM identifiers, hybrid combiners and signature scheme values. If a dependency changes, the MLS registrations may need to change with it. Teams should pin the exact draft revision in experiments and avoid treating provisional identifiers as stable production wire values.

Pitfall

Last call is not release evidence

An IETF last call asks whether a document is ready to advance. It is not final publication, IANA registration, an implementation release or a deployment receipt. Verify the eventual RFC, assigned identifiers, library release notes and product telemetry separately.

What implementers can test now

A controlled prototype can still produce useful migration evidence. Pin draft 08, a specific MLS implementation commit and explicit algorithm dependencies. Generate groups with each candidate suite, capture complete serialized messages, then replay negative cases. Useful failures include an unsupported suite, a mismatched credential signature algorithm, a malformed ML-KEM encapsulation key, a corrupted ciphertext and a member update signed under the wrong credential.

Do not test this with real conversations or production identity keys. Synthetic groups are enough to measure compatibility and message growth. Keep the fixtures so a later draft or library release can be compared byte for byte. The PQC migration roadmap treats this as an experiment and inventory phase, not an authorization to change a live group protocol.

What to wait for before deployment

  • The final RFC text and permanent IANA cipher-suite values.
  • Released library support that names the final specification rather than a development branch.
  • Interoperability across at least two independent implementations.
  • Documented handling for clients that cannot process the selected suite.
  • Credential issuance and rotation that match the chosen signature property.
  • Operational limits for message size, CPU, memory and large-group commit processing.

For a post-quantum migration, the immediate task is to map where group messaging is implemented and decide which security property is required. For an audit, verify the installed library, actual wire suite and credential algorithm. A roadmap that says only ML-KEM misses the distinction the draft makes explicit.

References

Get started

Discuss your security needs

Tell us what you need to assess, certify, set up, or teach. We’ll help you scope the work.