AI URI Scheme – Internet-Draft
Key topics
The Internet-Draft proposing an "AI URI Scheme" has sparked intense debate, with many commenters scratching their heads over the document's purpose and legitimacy. Some, like eob, are struggling to understand its practical applications, while others, such as progval, are raising red flags about the author's opaque background and affiliation with the "Artificial Intelligence Internet Foundation". As the discussion unfolds, it becomes clear that the draft's vague wording and lack of concrete examples have left many questioning its validity, with some, like wongarsu, pointing out that crucial details seem to be missing. Amidst the skepticism, a few commenters, like lgrapenthin, are digging up newer versions of the draft, shedding more light on the proposal's evolution.
Snapshot generated from the HN discussion
Discussion Activity
Active discussionFirst comment
2d
Peak period
12
36-39h
Avg / period
12
Based on 12 loaded comments
Key moments
- 01Story posted
Dec 15, 2025 at 2:28 AM EST
20 days ago
Step 01 - 02First comment
Dec 16, 2025 at 3:22 PM EST
2d after posting
Step 02 - 03Peak activity
12 comments in 36-39h
Hottest window of the conversation
Step 03 - 04Latest activity
Dec 16, 2025 at 4:24 PM EST
18 days ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
So it seems to be a random person cosplaying as a spec author. Or possibly trying to have something impressive on their résumé.
I don't understand what problem this is trying to solve though.
I don't get how this is better than an HTTP API (especially since payloads are just UTF8 json), and that's entirely down to the document not telling us anything of substance. I get it's "experimental", but there isn't much of an experiment being described here apart from a different message frame that allows us to leave out the http headers and add a signature (while apparently using the assumption that each ip only hosts one AIIP service)
I get very strong "E = mc^2 + AI" vibes from it, just shoehorning the coveted letters everywhere
Does the IETF have a red fat button on their desks, like the one often seen on variety shows, to instantly disqualify low effort submissions? You can infer that the document (and the interaction between the culprit and the IETF) was half-assed with the help of AI: https://mailarchive.ietf.org/arch/msg/art/ss-g3OHTtwHwyBDl0c....
In general, interoperability and user choice are really important for us to get right as the community of people building AI platforms...
Have others reading this document been able to map it onto their work?
As a specific example:
> ai://bank/service/payments?amount=10¤cy=USD
I'm not sure what this is representing here. Is it a way to encode a clickable link to chat with `bank` about `service/payments` with a few additional args attached?
AGI?
I was not expecting such ambiguous and inaccurate wording from IETF. Why "ai" and not, how it was traditionally called on the web, "robots"?
And of course this does not make any sense since vast majority of HTTP traffic is already autonomous.