What Is NostradamusGPT?
NostradamusGPT is a prototype prediction engine built directly on the Universal Law of Refinement (ULR) described in The Reason for Everything. It isn’t a generic AI model with a fancy name—it’s a live test of what happens when you actually let ULR drive a forecasting tool.
ULR starts from two ideas:
- The Will to Power – a universal drive that pushes every system to refine itself over time.
- The Asymptrex – the limit each system moves toward but can never fully reach.
If refinement is lawful and not random, then in principle the future path of any refining system is not arbitrary. It follows a curve shaped by the accumulated pushes and pulls of the Will to Power acting inside that system.
The Core Bet Behind NostradamusGPT
NostradamusGPT exists to explore one specific claim:
If we can model every meaningful iteration of the Will to Power inside a system, we can model that system’s future with remarkable accuracy.
In practice, that means treating things like user growth, skills, habits, and long-term projects as refinement curves rather than disconnected events. NostradamusGPT takes the ULR equations—the same refinement curves used in the book—and uses them to generate forward projections based on limited real-world input.
Instead of saying “anything could happen,” the engine assumes:
- The system is refining toward some Asymptrex (a realistic upper or lower limit).
- The Will to Power is what’s driving it there.
- Once you capture enough of that motion, the shape of the future becomes constrained.
Why This Prototype Exists
NostradamusGPT is not meant to be a polished commercial product. It’s just a fun public experiment:
- Can a law of refinement move from theory to practice?
- When we give a system just a little bit of data plus a plausible limit, does the ULR-based curve resemble what actually happens over time?
By letting people plug in their own data and see a predicted refinement path, NostradamusGPT turns ULR from an abstract framework into something you test and play with.
If the predictions consistently line up with reality—even roughly—then the implication is unsettling but powerful:
Parts of the future may be modelable, not just in hindsight, but in advance—whenever the iterations of the Will to Power inside a system can be captured and expressed through the Universal Law of Refinement.
NostradamusGPT is hosted by OpenAI, you can try it here: https://chatgpt.com/g/g-67f70060d6188191aff8759de386691b-nostradamus-gpt
Have fun.