News for hackers. Stuff that matters.

A comment on Anthropic: Progress on provable inference for large language models. Parent comment

Re:Anthropic: Progress on provable inference for large langu... by yorwba on Wednesday November 04, 2026 @03:44AM (#50362898)
If you trust the token generator and just worry about tampering in transit, signing the tokens after the fact is a fine solution.

This work is about the case where you trust nobody and therefore require even the token generator to do extra bookkeeping to prove that it's not taking any shortcuts.

[ Parent | on HN ]
Re:Anthropic: Progress on provable inference for large langu... by Tepix on Wednesday November 04, 2026 @05:04AM (#50363413)
The scenario is that I access Fable via openrouter and I don't want to be rerouted to an inferior model. If the vendor (Anthropic) sends a signature at the end of each response, that would solve the issue, wouldn't it?
[ Parent | on HN ]