As GPU data transfer speeds escalate, traditional electricity-based communication between nearby chips faces physical limitations. The industry is shifting to optics (light) for this "scale-up" networking. Nvidia is likely to acquire a company like IR Labs to secure this photonic interconnect technology, crucial for future chip architectures.

The race for AI performance has shifted from optimizing compute on a single chip to solving system-level connectivity. According to chip infrastructure startup Eliyan, efficiently linking chiplets, chips, and racks is now the primary bottleneck for building next-generation AI systems and maximizing compute utilization.

The core challenge of modern AI data centers is managing extreme power density and heat as compute gets packed tighter. The industry is spending trillions to solve the physics problem of moving data over millimeters within chips and racks, a radical shift from traditional data transmission.

With Moore's Law dead, Shkreli argues the future of computing lies in photonics. Using light for matrix multiplication (MATMOLs) offers a theoretical 1,000x to 1,000,000x performance gain, making it the necessary next frontier despite major technical hurdles.

While NVIDIA's GPUs have been the primary AI constraint, the bottleneck is now moving to other essential subsystems. Memory, networking interconnects, and power management are emerging as the next critical choke points, signaling a new wave of investment opportunities in the hardware stack beyond core compute.

Martin Shkreli makes a case for photonic computing—using light instead of electrons—as the next major paradigm in AI hardware. He argues that because matrix multiplications (95% of a GPU's job) are a natural function of light interference, photonic chips could offer an "insane speedup" with O-of-one complexity, making them a potential successor to GPUs.

SambaNova's architecture is optimized for inference by treating it as a data movement challenge rather than a raw compute problem. By designing for efficient data flow and communication between memory and compute units, they achieve 5-10x performance improvements over traditional GPUs.

With Moore's Law over, computing progress now depends on networking vast numbers of chips. Lightmatter's photonic interconnects overcome the distance limits of copper cables, allowing thousands of GPUs kilometers apart to function as a single, cohesive supercomputer. This creates a new scaling vector for AI performance.

The transition to AI workloads necessitates a total data center redesign. The physics of AI compute—extreme power density, heat, and bandwidth needs—are forcing a shift from transmitting data kilometers to millimeters. This creates opportunities across the entire physical infrastructure layer.