The Federal Trade Commission (FTC) has approved Elon Musk’s acquisition of Mesh Optical Technologies, a startup founded by former SpaceX engineers developing optical data center communication links. The clearance allows Musk to move forward with the deal for the Los Angeles-based company, which came out of stealth in February with a $50 million Series A led by Thrive Capital.
Why Musk wants Mesh Optical Technologies
To put a bit of context, Mesh was founded by Travis Brashears, Cameron Ramos, and Serena Grown-Haeberli, who previously worked on optical communications for Starlink satellites. While designing new satellites, they identified bottlenecks in optical transceivers (devices that convert optical signals into electrical data for computers).
For artificial intelligence (AI) data centers, these transceivers are critical: a million-Graphics Processing Unit (GPU) cluster requires four to five million transceivers. To this point, Mesh’s redesign removes one power-hungry component, potentially reducing GPU cluster energy consumption by 3 percent to 5 percent, resulting in significant savings for hyperscale operators.
American-made AI infrastructure
So far, the optical transceiver market is dominated by Chinese suppliers. Mesh Optical Technologies sees an opportunity to build a U.S.-based supply chain ahead of potential trade restrictions. The company aims to produce 1,000 units per day by year-end and qualify for bulk orders in 2027.
Thrive Capital Partner Philip Clark commented: “If AI is the most important technology in several generations, to have critical parts of AI data center capex run through misaligned/competitive countries is a problem.”
The orbit-to-AI pipeline: Mesh as the missing link
The Mesh acquisition fits into a much larger strategic vision Musk has been laying out since SpaceX’s initial public offering (IPO). Just weeks ago, Musk revealed SpaceX’s plan to build orbital AI data centers, constellations of AI satellites that would function as compute nodes in space, powered by solar energy and cooled by the vacuum of orbit.
The proposed AI1 satellite would deliver around 120-150 kilowatts (kW) of computing power, using Nvidia chips and technology derived from Starlink V3 satellites. SpaceX plans to launch millions of these orbital data centers, with a dedicated “Gigasat” factory in Texas expected to reach meaningful production by the end of 2026.
And this is where Mesh becomes critical. Orbital data centers require high-speed, low-latency optical links to communicate with each other and with ground stations. Mesh’s transceivers (designed to reduce power consumption by 3 to 5 percent and built on U.S. soil) could be the optical backbone of this space-based AI infrastructure.
For instance, by acquiring Mesh, Musk is not just buying a terrestrial data center supplier, but he is securing the optical plumbing for his orbital compute empire. The irony? Mesh’s founders built their technology for Starlink satellites. Now it’s coming full circle, back into space, powering the next generation of orbital AI, if they get to work, of course, because there’s still work to do around, and concerns about whether they really can do the job.
