The Quantum Antenna: Inside the Thirty-Year Scientific Journey That Rethinks How Machines Capture Light

For thirty years, Dr. Wolf Kohn has been chasing the same question, largely out of view: what if we’ve been thinking about computation the wrong way? Today, he pursues that question as chief scientist of Project LightShift, a research company he leads alongside CEO Nadab Akhtar, focused on rethinking how machines capture and process information from the physical world, starting with the antenna.

The question first appeared in print in 1996, when Kohn co-authored a Quantum Wave Processor Research Report through Cornell University’s Mathematical Sciences Institute, sponsored by the U.S. Army Research Office. Instead of treating computation as binary information moving through electronic circuits, the paper proposed representing computation as quantum states propagating across a programmable physical lattice. At the time, the idea was almost entirely theoretical: the materials, fabrication methods, and computational tools needed to test it didn’t yet exist.

Nearly thirty years later, Kohn believes the world has finally caught up. “People think I’ve spent my career working on quantum computing,” he says. “I haven’t. I’ve spent my career trying to understand how nature processes information.” That distinction shapes everything he and Akhtar do at Project LightShift.

Today’s technology conversation is dominated by AI: bigger models, faster chips, more computing power. But Kohn argues intelligence doesn’t start inside the processor. It starts the moment a machine observes the physical world. Akhtar, who often translates decades of theoretical physics into language engineers and investors can use, puts it plainly: “AI only works with the information you give it. If the observation itself is incomplete, no algorithm can recreate what was lost.”

That idea is the foundation of Project LightShift’s work on antenna design. Rather than building an antenna as a passive component that simply receives electromagnetic signals, the team is exploring an architecture that captures energy across multiple wavelengths, separates photons by frequency, converts them toward common frequencies, and synchronizes the resulting signals using quantum phase-control techniques, all before conventional downstream processing even begins. The company describes this in its technical materials as a quantum photonic antenna architecture, distinct from a conventional radio-frequency antenna. Whether it fulfills that ambition is still an open engineering question, but it reflects a broader shift: increasingly, the bottleneck in advanced systems isn’t computation. It’s perception.

Nowhere is that clearer than in defense. Military advantage used to be measured in fleet size, aircraft speed, or weapons power. Today, many analysts argue the deciding factor is informational: whoever observes, interprets, and reacts first gains the initiative. Recent conflicts have reinforced that shift. Small autonomous drones, often costing a fraction of traditional platforms, now handle reconnaissance and precision tasks once reserved for expensive aircraft. But those drones are only as good as the information feeding them.

“Autonomy doesn’t start with artificial intelligence,” Akhtar says. “It starts with perception.” Every autonomous system depends on reading an environment that’s growing more crowded and contested by the day. Cameras, radar, infrared, radio, and electronic sensors each contribute a piece of the picture, but simply gathering more data isn’t the answer; modern systems already produce more than they can use. The real challenge is making sense of it fast enough to matter, a problem that only intensifies as drones move from solo aircraft to coordinated fleets. “The future probably isn’t one extraordinary drone,” Akhtar says. “It’s thousands of capable systems working together.” That kind of coordination places enormous demands on sensing technology, since nothing downstream, including object recognition, target classification, and fleet coordination, can work if the initial capture is flawed. As Kohn puts it: “If you lose information at the beginning, you can’t recover it later.”

Revisiting the 1996 Quantum Wave Processor report today, its core ideas hold up: computation expressed through wave propagation, programmable physical systems, Hamiltonian control, and spectral analysis rather than digital switching. “The ideas have been surprisingly consistent,” Akhtar says. “What’s changed is the world around them.” Nanofabrication has advanced, photonic engineering has matured, and AI now offers tools for analyzing complex physical systems that didn’t exist three decades ago.

Kohn is careful not to overstate where things stand. “This isn’t about saying we’ve solved everything,” he says. “Science doesn’t work that way. Every answer creates new questions.” Akhtar adds that this kind of patience is rare in an industry obsessed with overnight breakthroughs: “We celebrate breakthroughs, but breakthroughs are usually the visible result of decades of work that almost nobody noticed.”

That framing sets the quantum antenna apart from typical tech narratives. It isn’t a sudden invention. It’s the latest chapter in a line of inquiry connecting mathematical theory, quantum physics, photonic engineering, and modern autonomous systems, one that spent decades waiting for manufacturing and computing to catch up to the theory. Asked what the goal has ultimately been, Kohn is unequivocal: “The goal has never been to predict the future. The goal has always been to understand it.”