Every month, we all contribute more to AI through writing, responding, conducting research, making reservations and even paying money. I now fully trust some of it. There are still parts of it that make me want to stay behind the wheel.
I would like to know where your line is. Which task would you gladly let AI complete on its own, and which would you never let it do without first verifying? There are no wrong replies; it could be personal or professional.
Since we are developing Clyro for just that space between "let it run" and "but keep it in check," I would love to know where people draw the line.
In the early LangGraph and GPT-4o days, we deployed a simple LangGraph agent for an internal tool. It worked well for about a week. Then a few colleagues told me some questions were taking several minutes, and it was failing most of the time.
I checked the trace logs and found the problem. The agent was stuck in a loop, calling the same function again and again until it filled up the context window. After 15 to 30 minutes of that, the request would just fail. Nothing was there to notice it had called the same function many times with no progress and stop it. This shows up everywhere.
Cursor just dropped "Origin," a GitHub competitor.
Cursor is moving way beyond "VS Code with AI." They just teased Origin, a Git-compatible forge whose landing page describes it as a tool built for "the agentic era," with code "moving faster than any infrastructure was built to handle." It's currently waitlist-only.
