Hello OAuth,

I noticed the upcoming interim agenda item on “Anthropic & OpenAI Agentic
Use Cases.”

This appears closely related to a protocol question I have been discussing
in the WG: where an AI system generates a concrete consequential action,
the generated action itself should not automatically constitute authority
to execute it.

My related Internet-Draft is:

https://datatracker.ietf.org/doc/draft-das-agentic-tool-binding/

The draft explores an act-bound authorization model in which the concrete
Candidate Act is first represented and validated, authority is then bound
specifically to that act, and the enforcement point verifies that the
operation being executed still corresponds to the authorized act.

I am not suggesting that this is necessarily the solution to the Anthropic
or OpenAI use cases, but I thought it may be useful context for the
upcoming discussion because it addresses the boundary between
model-generated arguments and execution authority.

Regards,

Sangam Das
Independent Inventor


On Thu, 3 Sep 2026 at 9:58 PM, IESG Secretary <[email protected]>
wrote:

> The Web Authorization Protocol (oauth) WG will hold a virtual interim
> meeting
> on 2026-09-28 from 13:00 to 14:00 America/Toronto (17:00 to 18:00 UTC).
>
> Agenda:
> Anthropic & OpenAI Agentic Use Cases
>
> Information about remote participation:
> https://meetings.conf.meetecho.com/interim/?session=35886
>
>
>
> --
> A calendar subscription for all oauth meetings is available at
> https://datatracker.ietf.org/meeting/upcoming.ics?show=oauth
>
> _______________________________________________
> OAuth mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
OAuth mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to