# Invoke Fibery AI Build Mode/Use User Permissions via Automations **URL:** \ **Category:** Ideas & Features\ **Tags:** ai\ **Created:** [March 20, 2026, 1:53pm UTC](https://community.fibery.io/t/invoke-fibery-ai-build-mode-use-user-permissions-via-automations/10609 "2026-03-20T13:53:28Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![RonMakesSystems](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/ronmakessystems/32/14122_2.png) [@RonMakesSystems](https://community.fibery.io/u/RonMakesSystems)\ **Post date:** [March 20, 2026, 1:53pm UTC](https://community.fibery.io/t/invoke-fibery-ai-build-mode-use-user-permissions-via-automations/10609/1 "2026-03-20T13:53:28Z")
Note that this would be with [AI mode in between "Ask" and "Build" mode](https://community.fibery.io/t/ai-mode-in-between-ask-and-build-mode/9518), not with build mode specifically, but with the permissions of the user set to be the invoker. This may be solved with MCP, but I’ve been struggling setting it up at the moment, and if it could all be done in Fibery, that would be pretty cool. Not sure if it aligns with the vision though… Imagine using an automation to trigger an AI which can make actions across your workspace. In the action setup, you have a formula field where you can set which user the ai should act on behalf of. There’s a question here on “who can set this up?”, “Do you need the user’s consent to have that AI action run on their behalf?”, “Should there be a different primitive for ‘AI Agent’ users, which anyone can create actions and automations for?” Just some thoughts ---
**Author:** ![Chr1sG](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/chr1sg/32/3941_2.png) [@Chr1sG](https://community.fibery.io/u/Chr1sG)\ **Post date:** [March 20, 2026, 2:04pm UTC](https://community.fibery.io/t/invoke-fibery-ai-build-mode-use-user-permissions-via-automations/10609/2 "2026-03-20T14:04:50Z")
> [@Granular AI Access Control (Space-level AI Permissions)](https://community.fibery.io/t/granular-ai-access-control-space-level-ai-permissions/10580/5): > > I think we should consider introducing the concept of AI agents as additional users in the workspace, potentially each under control of a human user. In other words, admins/architects can define what access each (human) user is allowed, and then any user can create ‘delegate AI’ users. By default, the AI users get the the same permissions as their human master, but the user also has the option of restricting them, e.g. making one of their AI delegates a read-only agent, making another limited …