Diodati Industries Letters
Letters

The Next Step Forward

November 2026

“You will see that I have tried to deduce the coefficient 8πν²/c³ in Planck’s Law independent of classical electrodynamics, only assuming that the ultimate elementary region in the phase-space has the content h³.”

Satyendra Nath Bose to Albert Einstein, 4 June 1924

Bose’s paper rebuilt the derivation of Planck’s law on a consistent foundation. Einstein called it “an important step forward”, translated it himself for publication, and carried the method further, from photons to other particles. The credit belongs to both, for their teamwork.

A century later, the physics they opened is becoming technology. Diodati Industries is being formed as a holding company to take the next step forward. This letter explains what we see, what we are building, and who we are.

What we see

AI, machine learning and quantum technologies are redefining our reality. They will change how we compute, how we communicate and how we sense the world around us.

Each of these technologies is powerful on its own. Together they are more powerful still. Quantum devices are beginning to compute, communicate and measure in ways that classical devices cannot, and they depend on conventional computers and AI to operate them. We believe the future belongs to systems that bring conventional computers, AI, machine learning and quantum technology together as one.

Signal and noise

Every one of these technologies meets the same obstacle: noise.

The information inside a quantum computer is fragile. Its errors cannot be avoided; they can only be detected and corrected, over and over, by a decoder that reads the machine's checks and decides what went wrong, fast enough to keep pace with the machine itself.

The most sensitive sensors face the same problem from the other side. The magnetic fields produced by the human brain are roughly a billion times weaker than the Earth's magnetic field. Quantum sensors can now measure them, but only when the signal is drawn out of a great deal of interference.

In both cases the answer is the same kind of work: repeated measurements, a precise model of the noise, and fast decoding that separates what matters from what does not. We choose difficult problems that have the greatest impact.

What we are building

The first company in this work is Bosonuum, which develops error-correction technology for photonic quantum hardware. Its product is decoding and simulation software for hardware whose main errors are detected losses, running on standard GPUs, and it is in development.

It is built for the architecture and error-correction teams at companies and laboratories building photonic, quantum-dot and neutral-atom quantum hardware. With it, they can learn, before building, how much photon loss their design can tolerate, and measure from their own test data the error rates their hardware achieves. Before producing any new result, its simulation reproduced four published error-correction benchmarks, each within a tolerance set in advance.

Where we are going

Sensing is a direction we are studying. The methods that protect a quantum computer from noise may also help a sensor recover a signal that is now lost in it. We will say more when there is something to show.

How we work

Einstein read a stranger’s claim and vouched for it in public. We hold our own work to that standard: a claim is checked before it is trusted. A prediction is tested against published results before we rely on it, and a result is reported with what it does and does not show. This is slower than announcing, and it is how dependable technology is made.

Who we are

We are drawn to difficult problems and to technology that does not yet exist, and we accept the patience that such work requires.

Elizabeth R. S. Burnim

Founder