The Session You Cannot take with you
This article discusses the loss of user control over AI session data as inference providers move toward non-portable, proprietary state formats. The authors argue for the necessity of session portability to ensure users retain ownership of their interactions.
Why it matters
Addresses critical concerns regarding data sovereignty and vendor lock-in within the rapidly evolving AI development ecosystem.
From: Earendil Engineering < rfc@earendil.com >
Subject: The Session You Cannot Take With You
The original promise of an inference API was wonderfully simple: send some input, receive some output. If you kept both, you had the conversation. You could inspect it, archive it, replay it, or give it to a different model.
That abstraction was never completely true. Prompt caches live on somebody else's GPUs, tokenization differs between models, and sampling is not reproducible (and quite intentionally so). But the semantic record of a session in the form of a transcript could still belong to the user. A transcript should contain the instructions, messages, tool calls and tool results. Another sufficiently capable model might not continue identically, but it could understand what happened and take over.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in