Figure · FIG:pilot/impairment-decision-flow

Held-and-used long-lived asset impairment decision flow

A bounded sequence separating the trigger, recoverability test, measurement, and later accounting.

Updated Aug 7, 2026 Review due Nov 7, 2026
For the bounded held-and-used model, an indicator leads to recoverability testing before loss measurement and prospective post-loss accounting.
Detailed visual description

The figure preserves the two-step ordering used in the project's held-and-used long-lived asset materials and explicitly warns that other impairment models may differ.

Review apparatus

Visual provenance and review

Technique
ai specified svg
Role
explanatory
Review state
candidate
Model source
google · google:gemini-2.5-pro
Rights
CC0-1.0 · Project-authored deterministic rendering of a reviewed model-assisted specification

Review notes

  • Google correctly emphasized the recoverability decision and no-loss branch.
  • Its raw proposal encoded nodes and edges as JSON strings and omitted the registered flow schema, so the central renderer rejected it.
  • The proposal motivated a finance-specific decision-flow renderer; the adapted spec removes example amounts and preserves the bounded-model warning.
Retained model-call description

scripts/figures/llm_spec.py constrained proposal using the complete C:long-lived-asset-impairment record; retained output records the learning question, rationale, proposed spec, and fact-check list.