I developed the following axioms after reading Pedro Domingos, There’s Only One Good Way to Regulate AI, on Medium – Im calling them, Domingos’ Five Axioms for Regulating AI

Domingos’ Five Axioms of AI Regulation

1. The Learning Axiom – AI systems constantly learn and change, making them fundamentally different from static technologies.
   
2. The Steering Axiom – Regulating AI is like steering a car – control the direction (metrics) rather than trying to map every possible route (behaviors).
   
3. The Balance Axiom – Every AI system should balance three interests: the company’s goals, society’s values, and the user’s wishes.
   
4. The Oversight Axiom – Only AI can effectively monitor AI, with humans guiding both sides.
   
5. The Control Axiom – Users must have a clear and simple way to tell AI systems what they want, just as a driver needs access to a steering wheel.