Live AI Operations
Writing on who can stop your AI agent, and how fast that stop really works.
I'm Myles Saulibio. This site collects some of my published posts on stops, credentials, and the work that keeps running after you think you've halted an agent.
Posts
Don't Ask the Agent. Ask Who Can Stop It.
Every model in one study said it would follow the rules. Some broke them anyway. Here's what to check instead.
Read the post →The month before Series A, and who may hit send
A short imagined run of a delegation audit when investor materials start to conflict.
Read the post →The Stop Has to Live on the Credential
My halt took 5.9 seconds. Then 41 side effects happened anyway.
Read the post →Demonstrated, Validated, Not tested
The three lines every stop test should end with, and why the last one matters most.
Read the post →A Revoke Is Not a Stop
“We can cut the agent's access, so we're covered" is one AI belief that isn't complicated. It's wrong, and here's the paper trail.
Read the post →You Already Knew How to Make the Bed
Helix 2.5, hidden notes, and why the scarce thing this week is not a better model
Read the post →Who Can Stop the System When Pressure Hits
AI safety fails at the decision boundary, not the performance layer
Read the post →Why AI’s Real Bottleneck Is Not Intelligence
The most useful idea in current AI discourse is not that models are getting smarter.
Read the post →Videos
Questions about the Brake Drill or your score? Sophie is an AI guide and can answer anytime, no booking. Talk with Sophie, or book 15 minutes with me.
Get new posts by email
Live AI Operations goes out on Substack. Free to subscribe.


