A physics-intelligence company

Intelligence from physics, not from probability.

NeoAmorfic builds an engine that reads the state of a system from the physics of its own record of motion. The engine is not trained and it does not forecast. It measures the current state of a structure and what has actually happened to it — order and disorder, the geometry of a trajectory, the coupling between its parts — and reports what that structure determines. Because the reading follows from the laws of physics, it cannot drift: the same record gives the same reading, on any machine.

Domains

Finance and biology, each with public evidence.

Evidence

Daily readings sealed to a hash chain since 27 July 2026.

Engine

QEIv18™, a physics engine certified on production in July 2026.

Company

NeoAmorfic Ltd, London. Incorporated November 2024.

Two families of machine intelligence now attract most of the world’s attention: language models, which learn the statistics of text and answer by sampling from them, and the physical intelligence of robotics, which learns to act in the world through a body. Both are learners, and both are probabilistic at their core. Neither is built to answer the question that sits underneath every decision made about a system in motion: what state is it actually in? Physics intelligence is our answer. It derives the state of a system from quantities that physics defines on the system’s own record, with no training and no probability. It does not generate and it does not act. It determines.

We say physics rather than deterministic, and the distinction matters. A rule can be deterministic and wrong; it will give the same wrong answer every time. A physical measurement reports a property the record actually has, and determinism is only the consequence: an answer that rests on the laws of physics cannot change between runs. Physics is also not a fixed toolbox. The engine began with entropy and the geometry of trajectories because those were the first measurements we could certify; it is built to admit further physical law as each new measurement is certified, and research now under way extends the same discipline to quantum systems.

Two domains, with evidence on the record

The engine is applied only where the data is a genuine record of a system over time. Two domains meet that test today, and both carry public evidence that anyone can check.

“Physics intelligence is a game-changer for any field where ‘close enough’ is catastrophic.”

NeoAmorfic is founder-led in the literal sense: the engine’s architecture, its verification rules and every product decision are Michael Anton J. Tupay’s. An inventor and serial founder trained in industrial design, in derivatives and in investment management, and an early adopter of every generation of computing including this one, he built the engine on one conviction — function follows form: read the structure of a thing and you know what it will do next — and to one standard: nothing is claimed that cannot be re-run.