AI·News & analysis
Google's first AI chips in orbit launch October 1
Project Suncatcher's prototype satellite, nicknamed MVP, carries Google TPUs into space on October 1 to test whether AI computing can work in orbit, where solar power is nearly constant.
Google launches MVP, its first Project Suncatcher satellite, on October 1, carrying four Trillium TPUs powered by a 1-kilowatt solar array.
The test checks whether the chips survive launch, radiation and cooling in orbit, running for just a few months. In the right orbit, solar panels stay lit almost constantly, up to eight times more productive than on Earth. Google plans a two-satellite laser-linked test in 2027, and expects nothing usefully operational for years.
What to know
- Google's first Project Suncatcher satellite, nicknamed MVP, launches October 1 on SpaceX's Transporter-18 rideshare mission.
- MVP is about the size of a refrigerator, built by satellite imagery firm Planet Labs and fitted with four of Google's Trillium TPUs, powered by a 1-kilowatt solar array.
- The test checks whether the chips survive launch, radiation and cooling in a vacuum, and will run for just a few months.
- Google plans a two-satellite test linked by lasers in 2027, and says nothing will be usefully operational for a few years.
Google is about to put its AI chips in space. Ars Technica reports that the first prototype satellite from Project Suncatcher, Google's research effort to run machine learning in orbit, launches October 1 on SpaceX's Transporter-18 rideshare mission.
What's on board
The satellite, nicknamed MVP and about the size of a refrigerator, was built by Earth-imaging company Planet Labs rather than from scratch, letting Google accelerate its timeline. It carries four of Google's Trillium tensor processing units (TPUs), roughly the computing power of a single server, running on a 1-kilowatt solar array, and will operate for just a few months.
What it will test
The goal is to learn whether AI hardware can survive and work in orbit:
- Launch: can the TPUs handle the heavy vibration of a rocket launch?
- Radiation: can the chips cope with years of radiation? Google has already tested them at UC Davis's Crocker Nuclear Laboratory.
- Heat: does the cooling system, built from heat pipes and radiators, keep the chips at a safe temperature in a vacuum?
Why space?
In the right orbit, solar panels get sunlight almost all the time. Google says that can make them up to eight times more productive than panels on the ground. The long-term idea is clusters of satellites connected by high-bandwidth lasers, working together as a data center. Power is becoming one of the biggest limits on AI data centers on Earth.
What's next
Google plans a test with two satellites linked by laser in 2027. It is also realistic about timing: the company says it doesn't expect anything usefully operational for several years.
Additional reporting: CNET, Engadget.
Key facts
- Project
- Google Project Suncatcher
- Hardware
- 4 Trillium TPUs, about one server
- Power
- 1-kilowatt solar array
- Launch date
- October 1, 2026, on SpaceX's Transporter-18 rideshare
- Partners
- Planet Labs builds the satellite; SpaceX launches it
- Next test
- Two laser-linked satellites in 2027
Got questions?
Quick answers, plain wordsWhat is Project Suncatcher?
It's Google's research project to run AI computing on satellites in orbit, where solar panels get near-constant sunlight.
When does the first Suncatcher satellite launch?
October 1, 2026, aboard SpaceX's Transporter-18 rideshare mission on a Falcon 9 rocket.
What's on the first satellite?
Four Google Trillium TPUs, about one server's worth of compute, powered by a 1-kilowatt solar array. The satellite, nicknamed MVP, was built by Planet Labs and is about the size of a refrigerator.
What will the test measure?
Whether the AI chips survive the launch, handle radiation in orbit, and stay cool in a vacuum using heat pipes and radiators.
When could AI data centers in space be real?
Google plans a two-satellite test linked by lasers in 2027, and says it doesn't expect anything usefully operational for several years.
Why put AI data centers in space?
In the right orbit, solar panels get sunlight almost all the time. Google says that makes them up to eight times more productive than panels on the ground, and power is a growing limit for AI.
SourcesArs Technica, CNET, Engadget
Topics and tagsGoogle, AI chips, Data centers, Space tech
Go deeper
Related stories
Google's new AI chips keep your data locked, even while they work on it
Google added two chips to its eighth-generation TPU family: TPU 8tx, which keeps data encrypted while AI runs on it, and TPU 8i Max, built for huge context windows.
Voters backed data center limits in dozens of midterm votes
Local measures to restrict AI data centers passed almost everywhere they appeared, and Democrats who promised pauses won Texas and Georgia. Elon Musk, who gave $474 million to Republicans, now faces a harder road for his AI build-out.
Google's AI spots signs of high blood pressure from a smartwatch
Google Research built an AI that finds signs of high blood pressure using only the pulse and motion sensors in a regular smartwatch. No cuff is needed to set it up.
More in brief
- YouTube is going back on the Billboard charts in 2027Nov 9
- Apple starts selling $50 packs for extra Apple Intelligence useNov 9
- Windows 11 tests a way to switch off the Sticky Keys popup for goodNov 9
- US regulators question whether Waymo's school bus fix workedNov 9
- NVIDIA unveils GR00T 2 and Isaac Lab 3.0 to teach robots delicate workNov 9
- Tesla launches its robotaxi service in San Antonio, its eighth cityNov 9