# Create a way to limit # of actions of an AI rule during testing/iteration phase to prevent infinite loops

**URL:** <https://forum.asana.com/t/create-a-way-to-limit-of-actions-of-an-ai-rule-during-testing-iteration-phase-to-prevent-infinite-loops/1126810>\
**Category:** Product Ideas\
**Tags:** agentic-work-management, ai-studio\
**Created:** [20 February 2026 20:14 UTC](https://forum.asana.com/t/create-a-way-to-limit-of-actions-of-an-ai-rule-during-testing-iteration-phase-to-prevent-infinite-loops/1126810 "2026-02-20T20:14:22Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![June\_LCG](https://luna1.co/8d76cd.png) [@June\_LCG](https://forum.asana.com/u/June_LCG)\
**Post date:** [20 February 2026 20:14 UTC](https://forum.asana.com/t/create-a-way-to-limit-of-actions-of-an-ai-rule-during-testing-iteration-phase-to-prevent-infinite-loops/1126810/1 "2026-02-20T20:14:22Z")

</div>

There needs to be a way to limit the number of actions that an AI automation can make during the testing/building phase to prevent infinite loops, like AIs endlessly commenting to each other, an easy mistake to make while iterating/testing. I burned through 10% of a month’s credits in 90 seconds yesterday, and deleted it immediately only because I had the billing/AI usage open in another browser window alongside…. Notice I said deleted - it wasn’t possible to halt it. Anyone else experiencing this?
