- From: Pierre Thierry <pierre@nothos.net>
- Date: Tue, 10 Jun 2025 21:20:49 +0200
- To: public-hydra@w3.org
Le 09/06/2025 à 19:48, Rigel a écrit : > One big hurdle, in my opinion, of REST is that the client needs to > understand the language of the server in order to use it. Prior to > LLMs, this was only possible if that language was coded into the > client. LLMs are very good at inferring this through a conversation, > which in this case are the RESTful state transitions driven by > hyperlinked and expressive resources. > > I think there might be an opportunity to put forth a new and better > standard for agent communication and cooperation. LLM agents are the > killer application for a Semantic Web powered by RESTful applications. Also, I think I've already seen severak news about AI agents either opening up a confused deputy attack, or accessing resources that their operator didn't want them to access. And REST lends itself very naturally to capability-based security, so that would be another interesting feature... > Shall we work on it? Sure! That leads to a question I was about to ask: what would be valuable to add to the current Hydra? So what would be valuable on the AI front? I'm actually wondering if there's anything AI-specific, actually, or if the goal shouldn't just be to have something complete and accesible enough for any consumer... Curiously, Pierre Thierry
Received on Tuesday, 10 June 2025 19:20:55 UTC