Google to Test AI Chips in Space With Project Suncatcher

Google is preparing to take a major step in its research into space-based artificial intelligence by testing its custom AI chips in low Earth orbit. The experimental mission is part of Project Suncatcher, Google's long-term research initiative exploring whether large-scale machine-learning infrastructure could eventually operate in space.

The first orbital test is designed to determine how Google's Tensor Processing Units (TPUs) perform under the harsh conditions of space, including launch vibration, radiation and extreme temperatures. Google says the mission is intended to collect real-world engineering data rather than demonstrate a fully operational orbital data center.

What Is Google's Project Suncatcher?

Google introduced Project Suncatcher in November 2025 as a research project investigating the possibility of scaling machine-learning computing infrastructure in space.

The concept involves using solar-powered satellites equipped with Google's TPU AI processors. These satellites could potentially operate as interconnected computing clusters in low Earth orbit, using the Sun as an abundant energy source.

According to Google, satellites in low Earth orbit can receive significantly more solar energy than solar systems on Earth because they can have access to sunlight for much longer periods. The company believes this could eventually provide an alternative approach to powering the growing computing requirements of artificial intelligence.

First Google AI Chip Test in Orbit

The upcoming mission represents the first in-orbit test of Google's Project Suncatcher hardware.

Google says the prototype satellite will be launched as part of SpaceX's Transporter-18 rideshare mission, developed in partnership with satellite company Planet. The satellite will carry Google TPU hardware into low Earth orbit, where engineers will monitor its performance in the real space environment.

The test will focus on several important questions:

• Can Google's AI chips survive the forces of launch?
• How will the TPUs perform when exposed to space radiation?
• Can the chips operate reliably in extreme temperatures?
• How effectively can the satellite remove heat from high-performance AI processors?
• What problems could affect future orbital AI computing systems?

The answers could help Google determine which technologies need further development before larger space-based AI systems become possible.

Why Does Google Want to Put AI Chips in Space?

Modern AI systems require enormous amounts of computing power. AI data centers on Earth consume substantial amounts of electricity and require sophisticated cooling infrastructure.

Project Suncatcher explores whether some of these challenges could eventually be addressed by moving computing infrastructure into orbit.

Solar power is one of the biggest attractions. Satellites can receive sunlight without the same day-night limitations experienced by solar installations on Earth. Google is therefore investigating whether networks of solar-powered satellites could provide energy for future AI workloads.

However, this does not mean Google is ready to replace conventional data centers with orbital facilities. The company describes Suncatcher as a long-term research effort, with significant engineering challenges still to be solved.

How Will Google Protect AI Chips From Space Radiation?

Radiation is one of the major challenges facing electronic equipment in orbit.

High-energy particles from the Sun and cosmic sources can interfere with electronic systems and cause errors. One example is a bit flip, in which radiation changes the value of a bit stored or processed by a computer.

Google has already conducted ground-based radiation testing of its TPUs at the University of California, Davis. The company reported that its Trillium TPUs performed well during testing and were able to withstand radiation exposure greater than the expected total dose for a five-year space mission.

The upcoming orbital mission will provide another level of testing because some space conditions cannot be perfectly replicated on Earth.

Cooling AI Chips in Space Is a Major Challenge

Keeping powerful AI processors cool is another major problem for orbital computing.

On Earth, data centers can use fans, air conditioning and other forms of conventional cooling. Space is different because there is no surrounding air in a vacuum to carry heat away.

Google is therefore researching alternative thermal-management technologies for Project Suncatcher. Its approach includes heat pipes and radiators designed to move heat away from the TPUs and release it into space.

Google has already tested its cooling approach in thermal-vacuum environments designed to simulate the conditions that spacecraft experience in orbit.

The first mission will help engineers understand how the cooling system performs under actual space conditions.

Google Plans Laser Links Between Future Satellites

Future versions of Project Suncatcher could require multiple satellites to work together as an interconnected AI computing network.

For that to happen, satellites would need extremely high-speed communication links. Google is researching laser-based communications that could allow satellites to exchange large amounts of data.

The company plans to launch two satellites in 2027 to test the high-bandwidth laser communication technology needed for future satellite clusters.

This would be an important step because AI workloads can require enormous amounts of data to move between processors.

Could Space Become the Next Location for AI Data Centers?

Project Suncatcher is part of a broader industry interest in space-based computing. Other companies are also exploring the possibility of putting computing hardware and data centers into orbit.

The potential advantages include access to solar energy and the possibility of using the cold environment of space for thermal management. However, space-based data centers also face significant obstacles, including launch costs, radiation protection, hardware reliability, maintenance difficulties and satellite manufacturing requirements.

Experts cited by Reuters have said that commercial-scale orbital AI infrastructure remains years away because of these technical and economic challenges.

What Happens Next?

Google's initial Project Suncatcher mission is primarily an engineering and research test. The company wants to collect real-world data, identify potential failure points and use those findings to improve future spacecraft and computing systems.

The next major milestone is expected in 2027, when Google plans to test two satellites and their high-bandwidth laser communication technology.

If the technology develops successfully, future generations of satellites could potentially carry larger numbers of AI processors and operate as connected computing clusters in orbit.

Final Thoughts

Google's Project Suncatcher represents an ambitious attempt to explore a new location for artificial intelligence computing. Instead of building every AI data center on Earth, Google is investigating whether solar-powered satellites could eventually provide computing infrastructure in low Earth orbit.

The first TPU test will not create a full orbital AI data center. Instead, it will answer fundamental questions about whether Google's AI hardware can survive launch, radiation, extreme temperatures and the unique cooling requirements of space.

For now, Project Suncatcher remains an experimental research project. But the upcoming tests could provide valuable information about whether space-based AI computing can become a practical technology in the future.