KAIST and Caltech Crack the Decades-Old Light Mystery Stalling Faster Internet
Two university labs have pinned down exactly how barium titanate controls and absorbs light, clearing the measurement uncertainty that kept a promising optical material out of commercial devices.

Researchers at KAIST and Caltech have precisely measured how barium titanate controls and absorbs light, resolving longstanding uncertainties that had slowed development of the material for ultra-fast optical communications. For decision-makers in networking hardware, semiconductors, and data centers, the result could shorten the path to optical devices that move data faster while wasting less light.
The secret was never in the material. It was in the measurements. Researchers at KAIST in South Korea and Caltech in the United States have precisely measured how barium titanate controls and absorbs light, resolving longstanding uncertainties that had slowed the material's development for decades. On paper that reads like a footnote in a physics journal. In practice it is closer to a permission slip. Barium titanate has long been one of the most promising candidates for ultra-fast optical communications, and what stood between it and commercial devices was not a shortage of promise. It was a shortage of reliable numbers.
Here is why that gap mattered. Optical communications move information as pulses of light, and the component that writes those pulses is a modulator. To design one, engineers need to know exactly how a material responds when voltage is applied, and how much light it swallows along the way. Without precise values, every device design becomes an educated guess, and every prototype carries a hidden penalty in lost signal. The KAIST and Caltech work pins down both behaviors in barium titanate: how it controls light and how it absorbs it. The reported payoff is direct - it could make it easier to build optical devices that transmit information faster while wasting less light.
Barium titanate is not a new discovery. It is a ferroelectric oxide, a class of material studied for decades that historically exhibits a strong response to electric fields, which is precisely the trait optical modulators want. That is what makes the long stall so frustrating to the field. The material kept surfacing in promising results and then failing to convert into dependable engineering. In materials science, that pattern usually traces back to characterization: different labs measure different values, the numbers refuse to agree, and device teams cannot build a production roadmap on a moving target. Settling the uncertainty is the unglamorous step that has to happen before anything ships.
For context, the photonics industry has spent years building around silicon because it plugs into existing chip manufacturing. Silicon is excellent at guiding light and poor at changing it on command. It lacks a strong native effect for switching light with voltage, so engineers historically pair it with other materials or other physical mechanisms, each carrying its own tradeoffs in speed, voltage, footprint, and loss. Barium titanate is one of the candidates that could carry that switching function. A material that both modulates light efficiently and absorbs little of it is the combination that lets you pack more optical channels into the same hardware and push each one harder.
That is where the business logic enters. Every optical link runs on a budget: the light you launch minus everything the components steal along the way. Absorption is the quiet line item that decides how far a signal travels, how much amplification it needs, and ultimately how much power the whole system burns. If barium titanate's absorption behavior is now measured precisely, engineers can stop padding designs with margin for the unknown and start optimizing. For the companies building optical transceivers, that translates into faster data rates per device, less waste, and a shorter path from lab result to product specification.
The timing matters because appetite for optical bandwidth has been climbing. Data centers, and especially the clusters used to train and serve large AI models, increasingly lean on optical connections between chips, boards, and racks, because copper runs out of reach and power headroom at high speeds. Faster optical devices that waste less light attack two constraints at once: throughput and energy. Any material advance that improves both deserves attention from anyone with capital tied to networking hardware, semiconductor manufacturing, or data center buildout.
Caveats belong here. A precise measurement is not a product. Turning barium titanate into commercial devices still requires integrating it with existing manufacturing, proving it holds up over years of operation, and showing it can be made at volume with consistent quality. The KAIST and Caltech result removes one specific blocker, the uncertainty around how the material behaves, and it does not remove the others. Historically, that is the step where many promising materials quietly die, and it is also the step that, once cleared, lets everything downstream move faster.
For executives and boards, the lesson is about where the bottleneck actually sits. In photonics, the ceiling is often not the physics of light but the reliability of the numbers used to design around it. A team that can trust its material data can commit to a roadmap. A team that cannot will keep hedging, and hedging is expensive in a market that rewards whoever ships first. Watch for replication of these measurements, for integration work that pairs barium titanate with mainstream photonic platforms, and for the first device makers to fold the new values into their designs. The mystery is closed. The race it unlocks is just starting.
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Science
Anthropic's I.P.O. Made Rank-and-File Researchers Worth Tens of Millions
Researchers who joined just a few years ago are now worth tens of millions, and the windfall resets the pay benchmark for every private A.I. lab still paying in paper.
Sea Lion Culling Strategy Fails: New Model Says Killing Them Won't Save Salmon
The latest research from AAAS overturn a long-held conservation assumption, forcing policymakers to rethink critical resource allocation in marine ecosystems.
Microplastics do more than pile up: they react with water to damage cells
Scientists have found that plastic particles can generate highly reactive molecules in water, a mechanism that could explain links to heart attacks, strokes, and neurological disorders.


