sprungWhat this page describes
How Sprung is being designed to work: permissions, approvals, history, escalation, and spending limits. These guide every feature we build for the beta.
Trust & control
Your AI team works for you, within limits you set.
Handing off work only makes sense if you can see it and stop it. These are the principles Sprung is being built on.
Five principles
Permissions are explicit
Each AI team member can only take the actions you’ve granted. If something isn’t on the list, the answer is no, and it comes to you.
Approvals where you want them
You choose which decisions need your sign-off: refunds, exceptions, estimates above a limit, policy changes. Sprung prepares them; you decide.
Every action is visible
What happened, who did it, and why is recorded on the customer and in your activity history. Nothing happens off the books.
Escalation comes with context
When something needs a person, you get the conversation, the relevant policy, and a suggested next step.
Spending stays within your budget
You set daily and monthly AI spending limits. Sprung tracks estimated spend against them and alerts you before you get close. Included usage and optional additional spending will be confirmed before a paid subscription.
What we’ll publish when it’s true
Sprung is in private beta. We’ll share specific security, data-handling, and hosting practices once they’re implemented and verified. We don’t claim certifications or guarantees we haven’t earned.
Read the privacy notice →