Coherent expands Texas photonics plant for AI infrastructure

Coherent broke ground on an expanded Sherman, Texas, manufacturing building for optical components used in AI data centers. NVIDIA’s Jensen Huang joined Coherent CEO Jim Anderson to highlight the role of photonics in scaling accelerated computing.

Coherent broke ground on an expanded manufacturing building in Sherman, Texas, aimed at scaling production of lasers, optical components and compound semiconductors used to connect AI systems. The company operates what it calls the world’s first 6-inch indium phosphide fab, producing wafers that help move data between chips, servers and data centers at optical speeds.

NVIDIA founder and CEO Jensen Huang joined Coherent CEO Jim Anderson, Sherman Mayor Shawn Temann and Texas economic development officials for the ceremony. The project is tied to CHIPS Act and Texas support, and builds on NVIDIA’s long-running relationship with Coherent, including a multiyear strategic partnership announced in March.

The expansion targets a growing bottleneck in large AI systems: connectivity. Huang said copper becomes inefficient when systems span racks and data centers, while silicon photonics can carry signals over longer distances with lower power demands. Coherent’s indium phosphide lasers, transceivers and pluggable optical modules are used in NVIDIA networking, including switches that rely on co-packaged optics.

Anderson said the Sherman site will support more than 550 direct jobs at full capacity, along with additional indirect employment. He framed the facility as part of a broader push to rebuild U.S. manufacturing capacity for advanced AI infrastructure.

58

Impact Score

AI regulators tighten rules as Anthropic passes OpenAI

European, UK and US authorities are moving toward more targeted oversight of high-risk systems, child safety and frontier models. Anthropic’s latest raise underscores how rapidly capital is concentrating around leading model developers.

Private evals become a strategic edge in AI

Satya Nadella’s push for internal learning loops points to a growing divide in enterprise AI. Fin and Cursor show how proprietary data, benchmarks, and usage traces can become durable advantages.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.