Glossary
Rung
One step on the Right Path: stop, identity, write door, pre-gates, evaluate, guardrail, autonomy, act, proof, metered.
Read the termGlossary · Term 05 of 39
Autonomy level
An autonomy level states how much an AI agent may do on its own, on a scale from A0 to A5. It is set per process in the declaration, checked at the autonomy rung of the Right Path, and recorded on every proof record, so anyone reading the record can see whether a person had to approve the action.
In one sentence: How much an agent may do alone, A0 to A5; recorded on every proof record.
01In more detail
What it means in practice
The autonomy level decides whether a person must say yes. Above the ceiling a write type is allowed to run at, a named approver who is not the asker sees the difference, the gates and the policy, and decides.
Earned autonomy uses a deterministic score, with no model involved. The level a write actually ran at is sealed into its record, so an auditor can see when a person was in the loop and when not.
02Where this shows up in elan1
Rung 6 AUTONOMY on the Right Path, and field F11 autonomy_level on every proof record.
04Questions
Autonomy level: common questions
A0 Observe reads only. A1 Draft produces a write intent that a person must submit; A2 Propose submits, and a person approves before the write; A3 Act reversible acts alone where a reversal plan exists; A4 Act bounded acts alone inside a declared blast radius and budget; A5 Act broad acts alone at tenant scope, and its declaration needs dual control.
On every proof record, in field F11 autonomy_level. It is checked at rung 6 AUTONOMY of the Right Path, so anyone reading the record can see whether a person had to approve the action. An actor without a grant resolves to A0 Observe.
Defined in the elan1 glossary · Last reviewed 10 Oct 2026. Definitions change only with a dated note.
// SEE IT
See the term in a governed action.
The walkthrough follows one write from intent to proof record, using the words in this glossary.