[saag] Re: Relay Attacks in Intra-handshake Attestation for Confidential Agentic AI Systems

[Eric Rescorla <[e-mail]>]

On Wed, Mar 4, 2026 at 1:49 PM Muhammad Usama Sardar <
[e-mail]> wrote:

> On 04.03.26 19:02, Eric Rescorla wrote:
>
> On Wed, Mar 4, 2026 at 9:59 AM Muhammad Usama Sardar <
> [e-mail]> wrote:
>
>> [ Already shared at relevant lists like SEAT, RATS, TLS, UFMRG; sharing
>> here for wider opinion/review ]
>>
> Please stop doing reposting this everywhere.
>
> I believe there are folks here who are not on other lists, and vice versa.
> I would very much value their perspective. There was a disclaimer upfront
> which folks on these list could safely ignore.
>

The whole purpose of having lists is so that people can subscribe to one
stream of messages and not others. Cross-posting to a bunch of
semi-irrelevant lists -- let alone reposting to them -- subverts this
purpose. The disclaimer doesn't make it better.


> It's not relevant to IETF activity.
>
> Ironically, the work to *exhaustively* explore all intra-handshake
> attestation options was requested by Paul, and Cocos AI is one of the major
> ones.
>
What's irrelevant here is not the technical questions about intra-handshake
attestation, but rather critiquing their activities elsewhere ("continuing
to make false
claims on social media").

-Ekr