Researchers have demonstrated single-shot tensor computing at the speed of light, marking a remarkable step toward next-generation AGI hardware powered by optical rather than electronic computation.
Tensor operations are a type of mathematical processing that underpins many modern technologies, especially artificial intelligence, but they go far beyond the basic math most people encounter. A useful comparison is the complex movements involved in rotating, slicing, or reorganizing a Rubik’s cube in several dimensions at once. Humans and traditional computers must break these steps into a sequence, while light can carry out all of them simultaneously.
In AI, tasks ranging from image recognition to language understanding depend heavily on tensor operations. As data volumes continue to grow, however, standard computing hardware such as GPUs is being pushed to its limits in speed, scalability, and energy use.
How Light Becomes a Calculator
Driven by the need for faster and more efficient computing, an international research team led by Dr. Yufeng Zhang of the Photonics Group at Aalto University’s Department of Electronics and Nanoengineering has developed a new way to carry out complex tensor calculations using a single pass of light. This technique enables single-shot tensor computing at the actual speed of light.
“Our method performs the same kinds of operations that today’s GPUs handle, like convolutions and attention layers, but does them all at the speed of light,” says Dr. Zhang. “Instead of relying on electronic circuits, we use the physical properties of light to perform many computations simultaneously.”
The team accomplished this by encoding digital information into the amplitude and phase of light waves, turning numerical values into measurable features of an optical field. As these structured light fields move, interact, and merge, they inherently perform mathematical processes such as matrix and tensor multiplications, which are essential to deep learning. Introducing multiple wavelengths allowed the researchers to expand this method so it could support even more advanced, higher-order tensor operations.
“Imagine you’re a customs officer who must inspect every parcel through multiple machines with different functions and then sort them into the right bins,” Zhang explains. “Normally, you’d process each parcel one by one. Our optical computing method merges all parcels and all machines together — we create multiple ‘optical hooks’ that connect each input to its correct output. With just one operation, one pass of light, all inspections and sorting happen instantly and in parallel.”
Passive, Efficient, and Ready for Integration
Another key advantage of this method is its simplicity. The optical operations occur passively as the light propagates, so no active control or electronic switching is needed during computation.
“This approach can be implemented on almost any optical platform,” says Professor Zhipei Sun, leader of Aalto University’s Photonics Group. ‘In the future, we plan to integrate this computational framework directly onto photonic chips, enabling light-based processors to perform complex AI tasks with extremely low power consumption.’
Ultimately, the goal is to deploy the method on the existing hardware or platforms established by major companies, says Zhang, who conservatively estimates the approach will be integrated to such platforms within 3-5 years.
“This will create a new generation of optical computing systems, significantly accelerating complex AI tasks across a myriad of fields,” he concludes.
Reference: “Direct tensor processing with coherent light” by Yufeng Zhang, Xiaobing Liu, Chenguang Yang, Jinlong Xiang, Hao Yan, Tianjiao Fu, Kaizhi Wang, Yikai Su, Zhipei Sun and Xuhan Guo, 14 November 2025, Nature Photonics.
DOI: 10.1038/s41566-025-01799-7
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
4 Comments
Memo 2511220803_Source 1. Reinterpretation Storytelling []
Source 1.
https://scitechdaily.com/scientists-just-made-ai-at-the-speed-of-light-a-reality/
1.
Scientists just made AI at the speed of light a reality.
Humans and conventional computers must perform tensor operations step by step, but light can perform all operations simultaneously.
Researchers have demonstrated single-shot tensor computing at the speed of light, representing a remarkable advance toward next-generation AGI hardware powered by optical, rather than electronic, computation.
1-1.
Tensor operations are a type of mathematical processing that underpins many modern technologies, especially artificial intelligence, but they go far beyond the basic mathematics most people encounter.
The complex operations of simultaneously rotating, cutting, and reconstructing a Rubik’s Cube in multiple dimensions are a good example. While humans and traditional computers must break these steps down sequentially, light can perform all of these processes simultaneously.
In AI, tasks ranging from image recognition to language understanding rely heavily on tensor operations. However, as data volumes continue to grow, standard computing hardware like GPUs are reaching their limits in terms of speed, scalability, and energy consumption.
2. How Light Becomes a Calculator
Driven by the need for faster and more efficient computing, an international research team led by Dr. Yufeng Zhang of the Photonics Group in the Department of Electrical and Nano Engineering at Aalto University has developed a novel method for performing complex tensor calculations using a single light wave. This technique enables single-shot tensor computing at virtually the speed of light.
“Our approach performs the same types of operations that today’s GPUs handle, such as convolutions and attention layers, but at the speed of light,” says Dr. Zhang. “Instead of relying on electronic circuits, we leverage the physical properties of light to perform multiple operations simultaneously.”
2-1. The research team achieved this by encoding digital information into the amplitude and phase of light waves, converting numerical values into measurable characteristics of the optical system.
What if we incorporated this encoding method into msbase? It would become a unique cryptogram that’s hard to imagine. Its data capacity would be infinite. Wow.
As these structured optical systems move, interact, and merge, mathematical operations essential for deep learning, such as matrix multiplication and tensor multiplication, are essentially performed. By introducing multiple wavelengths, the researchers were able to extend this method to support more advanced, higher-order tensor operations.
Multiplication of convolution matrices and tensors is the only complex way for msbase to propagate into electromagnetic waves. However, there are few shortcuts. It’s a maze, so even if you do, it’s difficult to understand.
If you need the speed of light, you need to obtain it from the msoss.mass.dark_matter value using oser.
>>>> If, by any chance, a cosmic AI needs the speed of light for faster computation than these, it must obtain it from qqcell. Dark energy provides that speed infinitely. Ugh.
>>>>> Of course, the human brain is always 10 billion light-years faster than AI due to the intertwined associative power of intuition and imagination. Ugh.
】
2-2.
_”Imagine you’re a customs officer, having to inspect every parcel and sort it into the appropriate box using multiple machines with different functions,” Jang explains. “Normally, you process each parcel one by one.
_But our optical computing method integrates all parcels and all machines into one. It creates multiple ‘optical hooks’ that connect each input to the correct output. With a single operation, a single pass of light, all inspection and sorting happens instantly and in parallel.”
3. Passive, Efficient, and Integration-Ready
_Another major advantage of this method is its simplicity. Optical computation occurs passively as light propagates, eliminating the need for active control or electronic switching during computation.
Professor Jifei Sun, leader of the Photonics Group at Aalto University, stated, “This approach can be implemented on virtually any optical platform.” He added, “In the future, we plan to integrate this computational framework directly into photonic chips, enabling light-based processors to perform complex AI tasks with extremely low power consumption.”
3-1.
Zhang conservatively estimates that this approach will be integrated into existing platforms within three to five years, with the ultimate goal being to apply it to existing hardware or platforms built by major companies.
He concluded, “This will usher in a new generation of optical computing systems that will significantly accelerate complex AI tasks in a variety of fields.”
If light is being used to model digital data, in multiple dimensions, and there is an endpoint to the process that can be converted back into digital format and displayed as a solution to a problem, that sounds to me like an analog computer is being employed as an intermediate step.
China is going kiIl everyone else in AI.
Just calm down.