Researchers at Chalmers University of Technology have developed a physics-informed artificial intelligence system capable of designing advanced optical materials up to ten times faster than traditional methods. By embedding the laws of physics directly into the AI model before training, the team has dramatically accelerated the process of predicting optical properties and engineering next-generation photonic devices. The breakthrough could speed the development of technologies ranging from lightweight camera and eyeglass lenses to advanced quantum computing components.

Figure 1. Laws of Physics.

Teaching AI the Laws of Nature

Unlike conventional machine-learning systems that must discover physical principles through massive datasets, the new AI starts with a built-in understanding of electromagnetism. This allows it to make accurate predictions while requiring far less training data and computational power. Researchers found that once trained, the system could analyze complex nanostructures and predict their optical behavior in milliseconds, delivering faster and more reliable results than previous approaches. Figure 1 shows laws of physics.

Accelerating Nanophotonics Research

The technology is being applied in the field of nanophotonics, where scientists manipulate light at scales smaller than its wavelength to create materials with extraordinary optical properties. Such materials could enable thinner, lighter, and more efficient optical components while supporting emerging quantum technologies. Working alongside quantum researchers, the team is also exploring how nanostructured photonic materials could help transmit information between future quantum computers using light-based communication systems.

Overcoming a Major Research Bottleneck

Traditionally, training AI models for optical design has required tens of thousands of computationally intensive simulations, with each simulation taking from several minutes to over an hour. Building a complete dataset could consume weeks of supercomputing time. By incorporating the fundamental equations of electromagnetism directly into the neural network, researchers reduced data-generation time from roughly 30 days to just three days while improving prediction accuracy.

Faster Path to Future Technologies

The new approach demonstrates how combining artificial intelligence with established physical laws can dramatically improve scientific discovery. By enabling rapid design and evaluation of advanced optical materials, the technology could accelerate innovation across photonics, communications, sensing, and quantum computing. The achievement highlights a growing trend in AI research: creating systems that not only learn from data but also understand the fundamental principles governing the physical world.

reference:

1. https://interestingengineering.com/innovation/new-super-brain-learns-laws-of-nature-to-fast-track-optical-component-design

Cite this article:

Keerthana S (2026), New AI System Deciphers Laws of Physics to Revolutionize Optical Engineering, AnaTechMaz, pp.963