DARPA launches PICASSO program to scale photonic integrated circuits
DARPA has launched a new research initiative aimed at transforming the scalability and system-level performance of photonic integrated circuits (PICs). The Photonic Integrated Circuit Architectures for Scalable System Objectives (PICASSO) program seeks novel circuit architectures that can enable very large-scale photonic circuits for applications including transceivers, photonic accelerators, sensing, and light detection and ranging (LiDAR).
According to DARPA, while photonic systems promise major gains in bandwidth, latency, and energy efficiency, current PIC-based solutions remain constrained to narrow functions and small component counts. This limitation has prevented photonic circuits from delivering clear system-level advantages over electronic architectures, particularly in compute and AI acceleration, where electronics still dominate overall performance.
The agency highlights two key technical barriers: signal attenuation and noise accumulation in long optical processing chains, and performance degradation caused by unwanted wave interactions such as scattering, back reflections, and spurious resonances. Addressing these challenges will require circuit-level and architectural innovation rather than incremental component improvements.
PICASSO aims to shift the focus of the photonics field from isolated components to holistic circuit- and system-level design, combining multiple mitigation strategies to enable massive scaling. DARPA expects the program to support new photonic architectures capable of outperforming electronic systems at the system level.
The program is scheduled to begin on 1 July 2026 and has an anticipated budget of $35 million. Phase one designs are expected to be taped out at photonic foundries within seven months, with system demonstrations targeted by month 18. Proposals are due by 6 March, with a proposers’ day planned for 16 January.
