Agents can waste money while trying to fix mistakes
AI agents running in can make an error and then spend more money trying to repair it. The cost grows when the agent repeats the same failed action or keeps retrying without making progress. This burns tokens while producing no useful result.
People running agents in real systems are comparing how often this retry loop happens, how expensive the worst cases have been, and how they stop runaway spending. Possible controls include , a manual , or simply accepting the bill.
Key points
- An agent can repeat a failed action while trying to recover from its own mistake.
- Retries with no progress can burn tokens and raise costs quickly.
- agents need for retry loops and runaway spending.
- , a manual , and repeat-action detection are practical .