Â鶹ÒùÔº

In a major leap for artificial intelligence (AI) and photonics, researchers at the University of California, Los Angeles (UCLA) have created optical generative models capable of producing novel images using the physics of light instead of conventional electronic computation.

in Nature, the work presents a new paradigm for generative AI that could dramatically reduce energy use while enabling scalable, high-performance content creation.

Generative models, including diffusion models and large language models, form the backbone of today's AI revolution. These systems can create realistic images, videos, and human-like text, but their rapid growth comes at a steep cost: escalating power demands, large carbon footprints, and increasingly complex hardware requirements. Running such models requires massive computational infrastructure, raising concerns about their long-term sustainability.

The UCLA team, led by Professor Aydogan Ozcan, has charted a different course. Instead of relying solely on digital computation, their system performs the generative process optically—harnessing the inherent parallelism and speed of light to produce images in a single pass. By doing so, the team addresses one of AI's greatest bottlenecks: balancing performance with efficiency.

The models integrate a shallow digital encoder with a free-space diffractive optical decoder, trained together as one system. Random noise is first processed into "optical generative seeds," which are projected onto a spatial light modulator and illuminated by laser light.

As this light propagates through the static, pre-optimized diffractive decoder, it produces images that statistically follow the target data distribution. Unlike digital diffusion models that require hundreds to thousands of iterative steps, this process achieves image generation in a snapshot, requiring no additional computation beyond the initial encoding through a shallow digital network and light illumination.

To validate their approach, the team demonstrated both numerical and experimental results across diverse datasets. The models generated new images of handwritten digits, fashion items, butterflies, human faces, and even artworks inspired by Vincent van Gogh.

The optically generated outputs were shown to be statistically comparable to those from advanced diffusion models, based on standard image quality metrics. They also produced multi-color images and high-resolution Van Gogh-style artworks, underscoring the creative range of the optical generative AI approach.

The researchers developed two frameworks: snapshot optical generative models, which produce new images in a single optical pass, and iterative optical generative models, which mimic digital diffusion to refine outputs over successive steps. This flexibility allows multiple tasks to be performed on the same optical hardware simply by updating the encoded seeds and the pre-trained diffractive decoder.

Beyond efficiency and versatility, the team showed that optical generative models can also provide built-in privacy and security. A single encoded phase pattern, generated from random noise, can be illuminated with different wavelengths, with each channel decoded only by its uniquely matched diffractive surface.

This creates secure, multiplexed image generation where the wavelength multiplexed content is inaccessible without the correct decoder—a capability not possible with standard free-space decoding due to cross-talk.

This physical "key-lock" mechanism ensures that unauthorized viewers cannot reconstruct the wavelength-multiplexed generated novel content delivered to individual authorized users, offering new opportunities for secure communication, anti-counterfeiting, and personalized content delivery.

The researchers also point to the potential of integrating optical generative models into wearable and portable devices, where compact, low-power designs are essential.

By replacing bulky modulators with nanofabricated passive surfaces or using integrated photonics, these models could be embedded in smart glasses, AR/VR headsets, or mobile platforms. Such implementations would enable real-time, on-the-go generative AI, bringing advanced content creation directly to users through wearable and portable systems.

The broader implications of this breakthrough are significant. Optical generative models could lower the energy footprint of AI at scale, making sustainable deployment possible while unlocking ultra-fast inference speeds. Potential applications extend across biomedical imaging, diagnostics, immersive media, and edge computing, where low-power, distributed AI is increasingly needed.

"Our work shows that optics can be harnessed to perform generative AI tasks at scale," said Professor Aydogan Ozcan, the study's senior author.

"By eliminating the need for heavy, iterative digital computation during inference, optical generative models open the door to snapshot, energy-efficient AI systems that could transform everyday technologies."

Looking ahead, the team envisions compact, low-cost optical generative devices enabled by advances in nanofabrication and photonic integration. Their ability to generate diverse outputs without digital bottlenecks may power future applications in secure communications, privacy-preserving content delivery, and distributed AI systems.

With this work, UCLA researchers have also pointed toward a sustainable and scalable future for machine creativity, signaling a convergence of photonics and artificial intelligence that could transform computing in the 21st century.

The authors of the work include Dr. Shiqi Chen, Yuhang Li, Yuntian Wang, Hanlong Chen, and Dr. Aydogan Ozcan, all from the UCLA Samueli School of Engineering.