Framework · Supplied arXiv:2609.11873v1 · 2026-09-10

B0–L5 autonomy framework

Which responsibilities in the improvement loop are controlled by AI, and which changes are inherited?

How to read B0–L5These are categories in an autonomy framework, not a single scalar score. B0 is the non-RSI reference level; different domains and systems can show different mechanisms.
B0

Non-RSI reference level

In-task AI improvement

Refine output

Changes refine an output during the current task or session. Future independent tasks inherit no accepted system update.

Criteria

  • No accepted persistent system change
  • Non-RSI reference level
View B0 assessments in the evidence ledger →
L1

RSI taxonomy level

Improvement execution autonomy

Execute improvements

Humans prescribe the target, update procedure and acceptance criteria. AI executes the procedure and accepted changes persist into later tasks or rounds.

Criteria

  • Human-defined what, how and success
  • Persistent accepted changes
View L1 assessments in the evidence ledger →
L2

RSI taxonomy level

Improvement strategy autonomy

Choose how to improve

AI diagnoses weaknesses and chooses interventions and experiments. Humans continue to set objectives, task boundaries and evaluation criteria.

Criteria

  • AI chooses improvement strategy
  • Objectives and acceptance remain external
View L2 assessments in the evidence ledger →
L3

RSI taxonomy level

Experience-acquisition autonomy

Choose what to learn

AI chooses or generates subsequent learning experience based on the evolving learner’s state. What it learns from changes as the learner changes.

Criteria

  • Learner-conditioned experience acquisition
  • Experience retained for later improvement
View L3 assessments in the evidence ledger →
L4

RSI taxonomy level

Environment adaptation autonomy

Adapt from deployment

Ongoing deployment or environmental interaction determines persistent changes in memory, skills, code, harnesses or parameters reused on later operational tasks.

Criteria

  • Persistent adaptation from operational feedback
  • High-level goals, access, evaluation and release may remain externally governed
View L4 assessments in the evidence ledger →
L5

RSI taxonomy level

Recursive inheritance autonomy

Improve the improvement mechanism

The procedure governing future improvement is itself revised, retained and invoked in subsequent improvement rounds. It may be an improver, search or research policy, evaluator or successor generator.

Criteria

  • Identify the revised mechanism
  • Verify creation, retention, inheritance and later invocation
  • Evaluate effectiveness separately from structural reuse
View L5 assessments in the evidence ledger →

Methodology and source authorship

The B0–L5 taxonomy is Duan et al.’s analytical lens, not a settled universal definition. Other taxonomies exist, including Weco’s outcome-focused four-level framework, which is not numerically interchangeable. Assignments depend on system boundary and inherited mechanism. Greater autonomy does not imply greater capability, safety, reliability or efficiency. Bounded mechanism demonstrations do not establish unconstrained autonomy. Capability results and autonomy assessments remain separate. Structural L5 requires a changed improvement mechanism to be created, retained, inherited and invoked later. Effective L5 additionally requires better subsequent improvements under comparable budgets and independent assessment. Repeated gains alone do not establish acceleration.

Extraction review: agent checked · Codex

Published snapshots

Duan et al. (Sep. 2026) assessment2026-09-10 · source author assessmentRSI Tracker: selected AI R&D evidence (Sep. 2026)2026-09-26 · tracker evidence synthesisRSI Tracker: selected evidence overview2026-09-26 · tracker evidence synthesis