Key Takeaways
- • CAS Star, a Chinese venture capital firm founded by Mi Lei and spun out of the state-run Chinese Academy of Sciences, has built a decade-long thesis in photonics that is now validating as AI infrastructure scales globally.
- • Over 200 of CAS Star's approximately 600 portfolio companies operate across the photonics sector, spanning sensing, communications, computing, storage, and display applications.
- • Photonics technology addresses a critical physical bottleneck in modern AI data centers: traditional copper interconnects generate excessive heat, cause signal loss, and create power consumption constraints that degrade chip performance as AI clusters scale to unprecedented sizes.
- • Optical links using light-based transmission instead of electricity have become central to next-generation data-center design, shifting investor and corporate focus toward photonics solutions as the foundational infrastructure for AI scaling.
- • Mi Lei's background as a doctorate holder in optics from the Xian Institute of Optics and Precision Mechanics positions CAS Star's photonics focus within deep technical expertise rather than speculative trend-chasing.
Summary
CAS Star, a venture capital firm founded by Mi Lei and emerging from China's state-run Academy of Sciences, is capitalizing on a decade-long investment thesis in photonics as artificial intelligence infrastructure reaches physical scaling limits. The firm has systematically built a portfolio of over 200 photonics-related companies across approximately 600 total holdings, covering sensing, communications, computing, storage, and display segments.
The timing of photonics' ascendance reflects a fundamental infrastructure constraint facing the AI industry. As machine learning models require training across tens of thousands of specialized computing chips operating in tandem, data transmission between these clusters has become a critical bottleneck. Traditional copper interconnects—the standard for decades—now face insurmountable physical limitations: electrical transmission through these cables generates excessive heat, causes signal degradation, and consumes disproportionate power as data volumes scale. These constraints directly reduce computational efficiency and increase operational costs in large-scale AI data centers.
Optical interconnects, which transmit data through light rather than electricity, circumvent these physical constraints by offering lower signal loss, reduced heat generation, and superior power efficiency. This fundamental advantage has transformed photonics from a niche technology into essential infrastructure for next-generation data-center architectures. The shift represents a validation of CAS Star's long-term strategic bet, as photonics has moved from a speculative thesis to a required solution for AI scaling.
Mi Lei's academic credentials—holding a doctorate in optics and maintaining ties to the Chinese Academy of Sciences—underscore the firm's foundation in rigorous technical expertise rather than trend-following investment patterns. This positioning has allowed CAS Star to identify and develop photonics companies during a decade when the sector lacked mainstream institutional attention, creating both portfolio depth and first-mover advantage as global investors and technology companies prioritize photonics infrastructure.
The timing of photonics' ascendance reflects a fundamental infrastructure constraint facing the AI industry. As machine learning models require training across tens of thousands of specialized computing chips operating in tandem, data transmission between these clusters has become a critical bottleneck. Traditional copper interconnects—the standard for decades—now face insurmountable physical limitations: electrical transmission through these cables generates excessive heat, causes signal degradation, and consumes disproportionate power as data volumes scale. These constraints directly reduce computational efficiency and increase operational costs in large-scale AI data centers.
Optical interconnects, which transmit data through light rather than electricity, circumvent these physical constraints by offering lower signal loss, reduced heat generation, and superior power efficiency. This fundamental advantage has transformed photonics from a niche technology into essential infrastructure for next-generation data-center architectures. The shift represents a validation of CAS Star's long-term strategic bet, as photonics has moved from a speculative thesis to a required solution for AI scaling.
Mi Lei's academic credentials—holding a doctorate in optics and maintaining ties to the Chinese Academy of Sciences—underscore the firm's foundation in rigorous technical expertise rather than trend-following investment patterns. This positioning has allowed CAS Star to identify and develop photonics companies during a decade when the sector lacked mainstream institutional attention, creating both portfolio depth and first-mover advantage as global investors and technology companies prioritize photonics infrastructure.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.