Search papers, labs, and topics across Lattice.
This review analyzes the potential of integrated photonics for AI acceleration, focusing on overcoming power, memory, and interconnect bottlenecks in traditional electronic computing. It introduces a bottleneck-driven taxonomy to identify operating regimes where photonics offers sustained benefits and emphasizes cross-layer co-design and workload-adaptive programmability for efficiency. The authors advocate for Electronic-Photonic Design Automation (EPDA) to enable closed-loop co-optimization and facilitate the transition from prototypes to scalable electronic-photonic ecosystems.
Photonics promises to break through the power, memory, and interconnect bottlenecks currently limiting AI's exponential growth, but realizing this potential requires a shift towards cross-layer co-design and automated electronic-photonic design tools.
The exponential growth of machine-intelligence workloads is colliding with the power, memory, and interconnect limits of the post-Moore era, motivating compute substrates that scale beyond transistor density alone. Integrated photonics is emerging as a candidate for artificial intelligence (AI) acceleration by exploiting optical bandwidth and parallelism to reshape data movement and computation. This review reframes photonic computing from a circuits-and-systems perspective, moving beyond building-block progress toward cross-layer system analysis and full-stack design automation. We synthesize recent advances through a bottleneck-driven taxonomy that delineates the operating regimes and scaling trends where photonics can deliver end-to-end sustained benefits. A central theme is cross-layer co-design and workload-adaptive programmability to sustain high efficiency and versatility across evolving application domains at scale. We further argue that Electronic-Photonic Design Automation (EPDA) will be pivotal, enabling closed-loop co-optimization across simulation, inverse design, system modeling, and physical implementation. By charting a roadmap from laboratory prototypes to scalable, reproducible electronic-photonic ecosystems, this review aims to guide the CAS community toward an automated, system-centric era of photonic machine intelligence.
