# AI First Principles

> AI First Principles is an open-source framework of 12 principles for organizations and teams responsible for operationalizing AI. It addresses the structural, ethical, and operational failures that emerge when AI is deployed without clear governance. The principles are grounded in lived organizational experience and validated by researchers, practitioners, and authors across AI, design, and organizational behavior. Licensed under CC BY 4.0 and free to share with attribution.

## Core Pages
- Home / Principles: https://www.aifirstprinciples.org/
  The complete list of 12 AI First Principles, each with a concise framing and a single actionable directive for teams implementing AI.
- The Treatise: https://www.aifirstprinciples.org/treatise
  Extended analysis of each principle including the hidden problem, cost of ignoring it, contrarian insight, solution framework, philosophical foundation, and research validation.
- AI Governance Framework: https://www.aifirstprinciples.org/ai-governance-framework
  A companion reference defining AI governance in the AI First Principles framework: ownership, failure visibility, transparency, user agency, and operational judgment.
- Operationalizing AI: https://www.aifirstprinciples.org/operationalizing-ai
  A companion reference for moving AI from experiment into real work without scaling inherited process dysfunction.
- AI Operating Model: https://www.aifirstprinciples.org/ai-operating-model
  A companion reference for defining how an organization owns, reviews, corrects, and changes AI systems.
- AI Constitution: https://www.aifirstprinciples.org/ai-constitution
  A companion reference for using the 12 AI First Principles as durable organizational constraints on AI decisions.
- AI Governance Checklist: https://www.aifirstprinciples.org/ai-governance-checklist
  A companion reference for reviewing ownership, failure signals, transparency, user agency, workflow discovery, ambiguity, and resource use before deployment.
- Contributors: https://www.aifirstprinciples.org/contributors
  Individuals and organizations whose work has shaped, extended, or built upon the AI First Principles, including contributors from NVIDIA, Google, Meta, Slack, ServiceNow, Adobe, and Harvard Business School. Each contributor has a dedicated profile page at /contributors/[slug] with biography, selected works, and their specific connection to the principles. Current contributor profiles include: Robb Wilson, Dave Thomas, Bryan Catanzaro, Cassie Kozyrkov, Kent Beck, Dan Goldin, Tim Wood, Sam Ransbotham, Anthony Franco, Cathy Pearl, Edward R. Tufte, Kara Swisher, Don Norman, Daniel Kahneman, Tim Brown, Josh Tyson, RJ Owen, Brian Solis, Laura Herman, Ben Goertzel, Catherine Joss, Daniel Lametti, Eric Ries, Lisa Feldman Barrett, Cathy O'Neil, Jeffrey Dastin, Timnit Gebru, I. Deborah Raji, Donna Haraway, Ian Goodfellow, Jeffrey K. Liker, Nicholas Diakopoulos, Richard McElreath, Dietrich Manzey, Zeynep Tufekci, Alex Rosenblat, Amy Edmondson, Mike Rother, John Shook, Rafael A. Calvo, Tom Beauchamp, Maurice Schweitzer, Rachel Croson, Kathleen Sutcliffe, Nassim Nicholas Taleb, Gary Klein, and Karen Hao.
- License: https://www.aifirstprinciples.org/license
  Full Creative Commons Attribution 4.0 International License terms and attribution guidance.

## Community
- Global Meetups: https://www.aifirstprinciples.org/meetups
  Directory of AI First Principles meetup chapters worldwide, with organizer information and event listings for practitioners applying the principles in their organizations.

## Resources
- GitHub Repository: https://github.com/aifirstprinciples/AI-First-Principles
  Open-source repository containing the full principles, treatise, practitioner prompt, and contribution guidelines.
- Practitioner Companion (ChatGPT): https://chatgpt.com/g/g-6890b95fdf6081919cc824e3360db1d1-ai-first-principles-practitioner-companion
  A ChatGPT-based companion tool for applying the AI First Principles in day-to-day AI implementation work.

## The 12 Principles (Summary)
1. AI Inherits Messiness - AI learns from people and shares their inconsistencies. Define what is prohibited over what is required.
2. AI Fails Silently - AI accumulates errors before patterns become visible. Build feedback loops over post-mortems.
3. People Own Objectives - Human accountability cannot be delegated to an algorithm. Name the owner.
4. Deception Destroys Trust - AI pretending to be human prevents calibrated expectations. Make AI obvious, not hidden.
5. Individuals First - AI industrializes manipulation at scale. Prioritize individual agency above efficiency or profit.
6. Build from User Experience - End users of failing systems are qualified to design system futures. Design from lived experience.
7. Discovery Before Disruption - Changing systems that are not understood creates unpredictable failures. Identify purpose before simplifying.
8. Ambiguity Is Wisdom - AI produces probabilities that demand judgment, not facts that replace it. Surface the probabilities.
9. Reveal the Invisible - Ignorance hidden in document theater must be surfaced. Pursue what is hard to explain.
10. Iterate Towards What Works - Grand plans commit to solutions without validating problems. Learn by doing, not planning.
11. Decompose Incrementally - Legacy systems are too brittle to automate wholesale. Dismantle complexity piece-by-piece.
12. Justify Resource Consumption - AI makes resource waste trivially easy. Optimize the ratio of value per resource spent.

## About
AI First Principles is an open-source project founded in 2025 by Anthony Franco (author of the WISER Method, Partner at FirstStrategy.ai). It is maintained at aifirstprinciples.org and hosted on GitHub under the Creative Commons Attribution 4.0 International License. The project serves AI practitioners, product managers, executives, and design teams navigating AI transformation. It has been adopted by organizations including OneReach.ai and covered by UX Mag.

## Contact
Email: info@aifirstprinciples.org
About: https://www.aifirstprinciples.org/about