Staera Semiconductor, a start-up specializing in optical connectivity for artificial intelligence systems, has emerged from stealth mode with its pioneering Lightwire technology. This innovation, aimed at improving data transfer efficiency in AI systems, has attracted US$4.6 million in pre-seed investment, led by Openspace Capital alongside investors such as SGInnovate, Black Kite Capital, 28 Holdings, In Group, and Decimus Technologies.
The growing demand for AI computing power has exposed a critical flaw in traditional electrical connections, which struggle under the heavy data loads required by modern processors. Staera’s Lightwire technology addresses this issue by embedding optical data connections directly into silicon chips, replacing conventional copper wiring. This approach not only enhances data transfer rates but is also more energy-efficient, a crucial consideration as data centers face sharp increases in electricity demand.
Staera’s innovation involves manufacturing the light source and other optical link components within commercial silicon wafers, ensuring compatibility with existing semiconductor fabrication facilities. This method integrates seamlessly into chips that connect processors and memory. Since 2016, Staera has been incorporating GaN LEDs into silicon chips, consistently achieving manufacturing yields between 98 and 99 percent.
Energy efficiency is central to Staera’s proposition. The company claims that Lightwire technology consumes less than one-tenth of the energy used by leading optical links currently available. It supports a transfer rate of 24 terabytes per second while using under 50 watts, compared to nearly one kilowatt required by existing co-packaged optics. This efficiency is particularly significant as the International Energy Agency forecasts a substantial rise in data center electricity consumption by 2030.
According to Staera’s CEO, Dr. Andrew Kim, the recent funding will facilitate the transition of Lightwire from research and development towards commercial deployment. The company plans to expand its US operations and establish partnerships with customers and supply chain entities, with a commercial release anticipated by 2030. Staera has already received commendations, having won the 3rd ECTC Start-up Competition in May 2026 and securing the runner-up position at the 2026 Lam Capital Venture Competition.
Hian Goh, Co-Founder of Openspace Capital, expressed optimism about Staera’s integrated optical approach, highlighting its potential to create more scalable and energy-efficient AI systems. As the industry moves towards more sustainable data-transfer methods, Staera’s Lightwire technology represents a promising advancement in AI infrastructure.
About Staera Semiconductor
Staera Semiconductor designs the technology that lets modern hardware scale, by carrying data on light instead of electricity. Its LED-based optical interconnect is fabricated inside a single chip, giving AI compute systems the density to move data from millimeters to meters without trading off size, energy or performance. Staera began as a spin-off from the Low Energy Electronic Systems (LEES) interdisciplinary research group at the Singapore-MIT Alliance for Research and Technology (SMART), MIT’s first research centre outside the United States and its largest international research endeavour.
For more information, visit https://www.staera.ai.
Amisha Shettigar
Blue Totem Communications
[email protected]
Legal Disclaimer:
The information contained in this article has been provided by independent third-party contributors, clients, or content partners. We do not independently verify the accuracy, completeness, legality, ownership, licensing, or reliability of submitted content, including text, images, videos, trademarks, or other media materials. The submitting party is solely responsible for ensuring that all content, including images and media assets, complies with applicable copyright, trademark, licensing, and intellectual property laws. We disclaim liability for any unauthorized use of copyrighted or proprietary materials by third parties. If you believe that any content published on this platform infringes your intellectual property rights, kindly contact the author above for prompt review and resolution.