The symptom is specific: aI costs rise because drafts are repeatedly regenerated. Start with the affected item and identify the decision or input that could produce this behavior.
Find the likely cause
The team retries vague requests instead of diagnosing defects.
Treat this as an explanation to verify against the actual work. Look at the input, the relevant decision, and the final result together. If the evidence does not support this diagnosis, investigate the mismatch before applying a convenient but unrelated fix.
Make the targeted correction
Classify the failure and revise the relevant instruction or input.
Cost control should reduce wasted work while preserving the quality needed for the final use.
Check that the repair worked
Check whether acceptance improves without increasing retry count.
Repeat the check on the final version that the reader or customer will encounter. An approved draft, a preview, and a published result can differ; the acceptance decision should concern the version people actually use.
Prevent the next related failure
A separate issue to watch for is this: low-cost drafts require extensive human rewriting. Compare end-to-end effort across approaches.
Monitor the useful outcome
Track total cost and human time per accepted item, including rework and failed attempts.
Keep a short record of the original symptom, the evidence behind the diagnosis, and the result of the acceptance check. That record makes the solution reusable when the same condition appears again, without assuming that every superficially similar problem has the same cause.