Bill Gates-backed startup unveils optical transistor GPU that could break Moore's Law and revolutionize AI computing forever

Neurophos, a startup supported by Bill Gates, has introduced the Tulkas T100, an optical processor designed to dramatically advance AI computing capabilities. This GPU operates at an impressive 56 gigahertz and supports a massive 768GB of high-bandwidth memory (HBM), enabling it to handle memory-intensive workloads with ease. The Tulkas T100 boasts a staggering 470 petaFLOPS of AI compute power, positioning it as a potential game-changer in the field of artificial intelligence and high-performance computing.

At the heart of the Tulkas T100 is the use of optical transistors, which are significantly smaller than traditional silicon photonics—by a factor of 10,000. This miniaturization allows for unprecedented transistor density and speed, potentially breaking the long-standing limitations imposed by Moore's Law. Moore's Law, which predicts the doubling of transistors on a chip approximately every two years, has been slowing down as silicon-based technology approaches physical and economic limits. Neurophos's optical transistor technology could thus represent a fundamental shift in how processors are designed and manufactured.

The dual reticle design of the Tulkas T100 integrates an enormous 768GB of HBM, a form of memory that offers extremely high bandwidth and low latency. This memory capacity is critical for AI workloads, which often require rapid access to vast datasets. By combining high-speed optical transistors with large-scale memory integration, the Tulkas T100 addresses both computational throughput and data handling challenges that have traditionally bottlenecked AI performance.

The implications of this technology extend beyond raw performance gains. Optical transistors consume less power and generate less heat compared to conventional silicon-based transistors, which could lead to more energy-efficient AI systems. This efficiency is crucial as AI models grow larger and more complex, demanding ever-increasing computational resources. Additionally, the compact size of optical transistors may enable more powerful processors to be integrated into smaller devices, expanding the possibilities for AI applications in edge computing and mobile platforms.

While the Tulkas T100 is still in the early stages of deployment, its potential to revolutionize AI computing is significant. By overcoming the physical limitations of silicon chips, Neurophos's optical processor could usher in a new era of faster, more efficient, and more capable AI hardware. This advancement aligns with growing industry demands for specialized processors that can handle the scale and complexity of modern AI workloads, potentially reshaping the competitive landscape of semiconductor manufacturing and AI research.

In summary, the Tulkas T100 represents a breakthrough in optical computing technology, combining ultra-fast operation, massive memory capacity, and energy efficiency. Supported by influential backers like Bill Gates, Neurophos is poised to challenge the dominance of traditional silicon-based GPUs and accelerate the evolution of AI hardware infrastructure.