Stateless

Stateless means carrying no information forward. The model is stateless across requests: each one resends the full context window, because the model has no way to see anything else. An agent is stateless across sessions by default: a new session starts empty, with no trace of prior ones.

The model itself is permanently stateless. Its parameters are frozen after training, and nothing you do at inference changes them. The model does not learn from your corrections, does not remember being told the same thing yesterday, and is not getting to know you, however much the conversation feels otherwise. The feeling of continuity within a session is manufactured by the harness, which keeps the transcript and re-sends it with every request. The model is not remembering the conversation, it is re-reading it.

The practical consequence: if you want something remembered across sessions, you have to write it down somewhere the agent will read it back. That is what AGENTS.md files, memory systems, and handoff artifacts are. When the agent keeps making a mistake you have corrected before, the question is not why it did not learn, it cannot, but where that correction should be written so every future session reads it.