In an update published on September 24, 2026, Google said the initial launch was scheduled for the following week as part of SpaceX’s Transporter-18 rideshare mission, developed in partnership with satellite company Planet. The prototype will evaluate how AI hardware handles the physical stresses of spaceflight and the radiation and temperature extremes encountered in low Earth orbit.
Project Suncatcher represents Google’s effort to investigate an alternative approach to expanding the computing infrastructure required for increasingly demanding AI workloads.
The initiative is based partly on the availability of solar energy in space. According to Google, satellites in low Earth orbit can access near-constant sunlight, potentially generating up to eight times more solar power than comparable systems on Earth.
Google is exploring whether that energy advantage could eventually support connected groups of satellites carrying specialized AI processors.
Under its longer-term concept, multiple satellite constellations could work together to process larger machine learning workloads, creating a distributed computing infrastructure in orbit.
However, Google emphasized that the project remains experimental. Its initial mission aims to determine whether existing AI hardware can operate reliably in space before the company attempts to develop more extensive computing infrastructure.
The first challenge is ensuring that TPUs can survive the stresses of rocket launches.
According to Google, a typical launch into low Earth orbit takes approximately 10 minutes, during which spacecraft can experience sustained acceleration of up to 10 times Earth’s gravitational force.
Individual electronic components can encounter even greater mechanical forces, potentially reaching 50 to 100 times Earth’s gravity.
To evaluate these conditions, Google’s engineering team conducted vibration tests that shook the prototype satellite along all three axes, simulating the frequencies and mechanical stresses associated with rocket launches.
The company reported that its hardware successfully withstood the vibration testing, providing an initial indication that the satellite and its computing equipment could survive launch conditions.
Another challenge involves exposure to radiation.
Outside the protection provided by Earth’s atmosphere, electronic components can be affected by cosmic rays and solar radiation. These conditions can damage hardware or cause computational errors, including changes to individual bits of information.
Google evaluated its Trillium TPUs at the University of California, Davis’s Crocker Nuclear Laboratory, exposing the processors to a proton beam while they executed AI workloads.
Engineers monitored the processors for errors and assessed how radiation exposure affected their ability to perform calculations.
Google reported that its initial tests showed the processors could withstand a total ionizing radiation dose exceeding what they would be expected to receive during a five-year space mission.
Although these results provide preliminary evidence of the hardware’s durability, the company intends to use the upcoming orbital mission to evaluate performance under actual space conditions.
Beyond launch survival and radiation resistance, Google is investigating how to manage the heat generated by AI processors in space.
Modern TPUs generate substantial heat in relatively small areas. Maintaining suitable operating temperatures is therefore essential to protecting the processors and sustaining computational performance.
On Earth, data centers rely on cooling systems that move heat away from servers and other equipment. However, the vacuum of space eliminates the ability to transfer heat through surrounding air.
Space-based computing infrastructure must instead depend on alternative thermal management systems that can transfer heat to radiators and release it into space.
Google is evaluating combinations of heat pipes and radiators as potential solutions for cooling its orbital computing hardware.
The company’s engineering team has already tested its proposed cooling technology in a thermal vacuum chamber designed to reproduce the temperature and vacuum conditions encountered in space.
The upcoming mission will provide an opportunity to examine how the cooling system performs in orbit and identify potential improvements for subsequent satellite designs.
Another engineering challenge is establishing high-speed connections between satellites.
Google’s longer-term plans envision satellite clusters in which each spacecraft carries dozens of TPU chips, allowing computing workloads to be distributed across multiple orbital platforms.
For this system to function, the satellites would need to exchange substantial amounts of information while maintaining precise knowledge of their positions relative to one another.
Google plans to use laser-based communication systems to establish these connections.
Although laser communications are already used in space, Google said its proposed application has different requirements from many existing satellite communication systems.
Rather than transmitting information across extremely long distances at relatively low bandwidth, Project Suncatcher requires very high-bandwidth communication over short distances between moving satellites.
Maintaining these connections will require precise positioning and tracking as the satellites travel through orbit.
Google plans to conduct its first orbital test of this technology in 2027, when it intends to place two satellites in orbit. The experiment will evaluate the laser communication capabilities required to support future interconnected satellite clusters.
The upcoming prototype mission is therefore an early step in a broader research program rather than the deployment of an operational space-based data center.
Its primary objectives are to gather information about hardware reliability, evaluate thermal management under actual orbital conditions, and identify engineering problems that must be addressed before larger systems can be developed.
Google expects to use the results to refine subsequent satellite designs and inform its next research milestones.
Project Suncatcher forms part of the company’s broader exploration of technologies that could eventually support the growing computational requirements of artificial intelligence.
While the feasibility of large-scale orbital AI infrastructure remains unproven, the initial launch will provide Google with its first opportunity to test its specialized AI processors in space.
