Human in the Loop vs On the LoopAn open-source guide to the oversight models an AI agent runs under: what human in the loop and human on the loop each decide, which action needs which, and the Kortix policy that enforces it.https://humanintheloopvshumanontheloop.com/2026-10-26T00:00:00ZHuman in the Loop vs On the LoopHuman in the loop vs human on the loop: open source oversighthttps://humanintheloopvshumanontheloop.com/2026-10-26T00:00:00ZIn the loop, a person decides before an action runs; on the loop, the system acts while a person watches. Open-source Kortix lets you set which, per action.In the loop, a person decides before the action runs. On the loop, the system acts while a person watches and can stop it. Kortix is the open-source AI Operating System, and the choice between them is a rule in a policy you write.Human in the Loop: An Open-Source Approval That Names the Callhttps://humanintheloopvshumanontheloop.com/human-in-the-loop2026-10-26T00:00:00ZHuman in the loop means a person decides before an action runs, and open-source Kortix binds each approval to one exact call and its arguments.Human in the loop means a person decides before an action runs; in open source Kortix that decision is a rule in kortix.yaml. Kortix is the open-source AI Operating System: agents, skills, memory and connectors live in a git repo you own, and every session runs on its own cloud computer.Human on the Loop: Open Source Supervision You Can Changehttps://humanintheloopvshumanontheloop.com/human-on-the-loop2026-10-26T00:00:00ZHuman on the loop keeps a person above a running agent. Open-source Kortix also gives that person the policy that decides which actions run.Human on the loop means a person supervises a running AI agent and can change the rule it works under; on open-source Kortix that rule is a file you own. Kortix gives a supervisor three levers: fix one outcome, change the rule, stop the run. Kortix is the open-source AI Operating System.The Human over the Loop Writes the Policy an Open-Source Agent Runs Underhttps://humanintheloopvshumanontheloop.com/human-over-the-loop2026-10-26T00:00:00ZHuman over the loop means a person writes the constraints an agent runs under; on open-source Kortix that policy is a file you review as a diff.Human over the loop means a person writes the constraints an agent runs under; in open source Kortix that constraint is a policy file in the project repo. Kortix is the open-source AI Operating System: agents, skills, memory and connectors live in a git repo you own, and every session runs on its own cloud computer.Ask or Allow: An Open-Source Test for One Agent Actionhttps://humanintheloopvshumanontheloop.com/how-to-choose2026-10-26T00:00:00ZSort one agent action into hold, supervise or allow by what survives a wrong call, then write the rule that enforces it in open-source Kortix.Open source puts the choice between human in the loop and human on the loop on each action: hold the call, supervise it, or let it run, and a rule you write enforces it. Kortix is the open-source AI Operating System.Open-source AI agent approval workflow: bind each approval to the callhttps://humanintheloopvshumanontheloop.com/approval-workflow2026-10-26T00:00:00ZAn open-source Kortix approval workflow is a policy file plus a held call: which action waits, what the approver reads, where the record lands.An open-source approval workflow for an AI agent is a rule, a held call, a named decider and a log, and the rule belongs in a file you review as a diff. Kortix is the open-source AI Operating System.