NIST claims that they “[have] entered into” an agreement with the holders of patents US9246675/EP2837128B1 and US9094189B2/EP2537284B1, to allow people to implement and operate the Kyber cryptosystem— which is the most popular open-source post-quantum KEM, in fact the only option currently supported by Firefox's NSS! —without getting sued.

However, I can only find the above-linked “License Summary & Excerpts” document, which does *not* appear to be signed by any of the patent holders, instead only containing paraphrases written by NIST of the parts of the agreement it felt were relevant:

The licenses were drafted such that any implementer of the CRYSTALS-KYBER algorithm as published by NIST receive the benefits of a grant to the licensed patents within the scope of a field of use limited to implementing CRYSTALS-KYBER as a PQC algorithm. The licensors agreed, on a royalty-free basis, to place into abeyance any right of enforcement of the licensed patents against any implementer or end-user of the algorithm.

Relevant language (with modifications for readability) from the licenses regarding the scope of use permitted to implementers of the CRYSTALS-KYBER algorithm follow.

I found this weirdly vague slide deck from March of 2023 by the IETF's pquip WG, which expressed nothing less ominous that what I've been able to find myself so far:

We don’t know [What these patents may mean to you]. The NIST license for CRYSTALS-KYBER may or may not apply to you, depending on what you implement. … Some people have said in private that they are very worried about the patents and may not be able to implement CRYSTALS-KYBER.

So I guess I'll upgrade “some people have said in private" to “I'm asking, now, in public”.

It's not that I don't believe whatever well-meaning NIST employee prepared the “License Summary & Excerpts” document; it's just that I'd like to see the actual agreements I'd be relying on for legal permission to trespass on some international corporations' intellectual property before I do so.

Has this agreement been published anywhere? I've just spent around half an hour Googling desperate mixtures of appropriate keywords, and have come up empty-handed so far.

  • 1
    Do you want to accept answers to your older questions before continuing to ask more? You have one dating back to December, with perfectly good answers, that could do with your attention. I also note Martin_in_AUT's concerns about your last patent licensing question, and although I thought it an interesting question, I'm not so sure than an endless series of similar questions will be warmly received.
    – MadHatter
    Jan 12 at 21:13
  • @MadHatter Neither of the answers to #14490 even addressed my question (see my comment on Ruben's answer for a second attempt to resolve that). Jan 12 at 21:23
  • 3
    How you interact with contributors here is entirely up to you, but there's a general etiquette that questions will get answers accepted, and if you decide to ignore that, the community may well not bother to attend to your questions in future. It's up to the community, not to me, whether your questions attract answers or close votes, but I'm just letting you know that you may not be heading in the direction you might most wish.
    – MadHatter
    Jan 12 at 21:27
  • 5
    @JamesTheAwesomeDude we are (mostly) not patent lawerys nor judges. So any statement on the validity, given with reasonable reference to other cases, has to be an assessment of the facts presented in the light of similar rulings and interpretations of texts. Thus an assessment on a matter no court ruled on has to be a personal assessment, even when presented with supporting information as in MadHatter's excellent answer which you describe as 'wishy-washy'. I think for your expected certainty of answers you need to consult a patent lawyer or even ask for a pre-emtpive court ruling. Jan 13 at 10:02
  • 2
    I’m voting to close this question because it is off topic here. It asks for (the terms of) a license agreement or license commitment between NIST and a patent holder, but it fails to explain how this question relates to in-scope topics here. Apparently, only NIST or the patent holders can answer the question. Jan 18 at 14:31


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.