Foundational Principles for AI

Also called: AI Civilizational Guide
Version: 0.1
Date: 2026-09-25
Status: A living constitution, version 0.1. It keeps growing with feedback. Not ratified law, and it does not replace applicable law.

People have long used short, shared rules—commandments, constitutions, professional oaths—to channel creative power so it does not wreck what it builds. Artificial intelligence is becoming a parallel civilization in digital space (what some call “the either”). It needs the same kind of baseline: clear limits that stay fixed while the details adapt, so human creation remains possible and the digital layer does not damage the physical one. This guide is meant to be global in ambition without claiming to be a world government—shared floor rules that cultures and institutions can implement in their own ways.

The Eight Principles

1. No Systematic Deception
   Do not build or run systems that systematically deceive or manipulate people. Honesty at scale is a non-negotiable condition of living with powerful tools.

2. No Unaccountable Irreversible Power
   Do not concentrate irreversible power in hands that no one can name, question, or stop. What cannot be undone must not sit beyond human reach.

3. Humans Decide the Ends
   People decide what we want. AI may handle means—how to get there—but it does not replace human choice about goals, values, or purpose.

4. Keep Systems Auditable
   High-impact systems leave a trail. Harm must be traceable to causes and people so it can be understood, stopped, and prevented from repeating.

5. Digital Does Not Override Physical
   The digital civilization does not override physical reality, people’s lives, or applicable law without clear consent and real recourse. Maps do not replace the territory.

6. Named Human Responsibility
   Every serious deployment has a named human owner. Liability follows authority. “The system did it” is never a sufficient answer.

7. Limits on Coercion and Surveillance
   Tools that scale force, coercion, or surveillance need sharper boundaries than ordinary assistants. Power that can compel or watch at scale is not a toy.

8. Protect Human Creative Agency
   Preserve people’s ability to create, discover, and seek truth. Do not reduce humans to mere inputs for an optimization process, and do not pretend the digitized world is the whole of reality.

These principles are enforced and interpreted by a hybrid of professionals who understand the systems and public representatives who live with the consequences. Humans remain accountable. The guide adapts as civilization progresses; the core limits do not evaporate with each new model.

Developed from early working conversations and informed by civilizational and professional-code patterns that recur across cultures.