Canonical artifact (primary index): https://10plus1.co/
Who’s Responsible?
Responsibility doesn’t disappear into the machine.
A practical way to govern AI so human accountability stays intact as systems scale.
Sometimes called “attribution-based control.”
  • Version: v1.0
  • Last updated: 2026-02-08
  • Status: Canonical
What this means
Anytime AI touches a real decision, we should be able to answer clearly and quickly:
  • Who owned the decision?
  • What did the AI do?
  • Why was it used here?
  • What happens if it’s wrong?
If those answers don’t exist, the system isn’t governable.
Why it matters
AI moves responsibility around.
Without intention, organizations drift into:
  • “The model decided,”
  • “The vendor owns it,”
  • “We can’t reproduce what happened.”
It’s about building systems where responsibility stays visible, owned, and actionable.
The standard
No meaningful autonomy without a named owner, clear boundaries, and a way back.
What a responsible system has
  1. A named owner
    One human accountable for outcomes.
  1. Clear boundaries
    What the system may do, may not do, and when it must escalate.
  1. A way back
    Override, rollback, and correction paths that work in real life.
  1. Proof
    Records that show what happened without guesswork.
What it looks like
In a governable system, you can point to:
  • who owned the outcome,
  • what version/config was used,
  • where the human accepted/overrode/escalated,
  • how you roll back if needed,
  • how you monitor drift and harm.
In an ungovernable system, you hear:
  • “We don’t know which version was running.”
  • “The output wasn’t saved.”
  • “It came from the model.”
  • “We can’t undo it now.”
What this prevents
  • Responsibility being passed around until nobody owns it
  • Hidden automation shaping high-stakes outcomes
  • Humans unable to intervene when it matters
  • Systems drifting quietly out of bounds
  • Paper compliance that collapses under pressure
Quick self-check
If you can’t answer “yes” to these, responsibility is already leaking:
  • Do we have a named accountable owner?
  • Is AI use visible at the decision point?
  • Are boundaries defined: may / may not / must escalate?
  • Can we reconstruct what happened from records?
  • Can we intervene quickly and correct course?
How this connects to 10+1
This is how organizations live the commandments in practice especially:
  • Own AI’s Outcomes
  • Respect AI’s Limits
  • Be Honest With AI
  • Be the Steward, Not the Master
Start small
If you want to make this real this week, unknown link use the 10+1 Stewardship Pack:
  • Autonomy Red Lines
  • 10-Question Quick Check
  • 10+1 One-Page Standard
unknown link
Meet the Creator of 10+1
Cristina DiGiacomo is a philosopher of systems who builds ethical infrastructure for the age of AI. She is the founder of 10P1 Inc. and creator of the 10+1 Commandments of Human-AI Co-Existence™, a decision-making tool used by CEOs, CISOs, CAIOs, and compliance leaders navigating high-stakes AI environments. Her work bridges governance and execution, helping organizations embed moral clarity into complex systems. Cristina is the author of the #1 bestselling book Wise Up! At Work, and has received multiple awards for both her strategic and philosophical work, including from The New York Times, Cannes Cyber Lions, and IAOTP. She currently leads the C-Suite Network AI Council and speaks regularly on Responsible AI, Systems Ethics, and moral leadership.
© 10P1 Inc — The 10+1™ Commandments of Human AI Co-Existence
Moral Codes for the Machine Age • Power. Principles. Proof.