Zuckerberg's Meta will beam sunlight from space to power AI data centers, solar-collecting satellites will orbit 22,000 miles above Earth — firm reserves 1 Gigawatt of orbital solar energy and 100 Gigawatt-hours of long-duration storage

satellite beams down sunlight from space
(Image credit: Meta)

Meta announced in a press release on Monday that it had begun plans to power its AI data centers with sunlight beamed from space. The company plans to place solar-collecting satellites in geosynchronous orbit roughly 22,000 miles above Earth to beam the sunlight to collectors on the Earth's surface. The plan is part of two new partnerships aimed at solving the rising energy demands of the company’s rapidly expanding AI infrastructure.

The company said it has reserved up to 1 Gigawatt of capacity from space solar startup Overview Energy, alongside another deal with energy storage firm Noon Energy for up to 1 GW / 100 Ggigawatt-hour of ultra-long-duration energy storage.

Latest Videos From

Meta’s partnership with Overview Energy attempts to solve that problem by moving solar collection into space — where the sun never sets.

Meta says it plans to deploy up to 1 GW of this “orbit-to-grid” energy system to support its data center operations. The first orbital demonstration is currently planned for 2028, with commercial delivery potentially beginning around 2030 if the technology works as intended.

satellite beams down sunlight from space

(Image credit: Meta)

The company’s second partnership focuses on another major weakness in renewable energy systems: storage.

While lithium-ion battery systems are effective for short-duration storage, they are generally not designed for sustained power delivery over multiple days. Meta says Noon Energy’s technology can provide more than 100 hours of energy storage using reversible solid oxide fuel cells and carbon-based storage systems. Under the agreement, Meta has reserved up to 1 GW / 100 GWh of storage capacity, with an initial 25 MW / 2.5 GWh pilot project expected in 2028.

Long-duration storage systems like these are becoming increasingly important as grids absorb larger amounts of renewable energy. Extended storage could allow excess renewable energy generated during peak production periods to remain available through prolonged low-generation periods, helping stabilize power availability for continuously operating AI infrastructure.

Meta says both partnerships are a part of its broader strategy to secure long-term energy supplies for future AI expansion. The company says it has already contracted more than 30 GW of clean, renewable energy and has also backed geothermal and nuclear energy projects, including agreements involving 7.7 GW of nuclear capacity.

Google Preferred Source

Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Etiido Uko
News Contributor

Etiido Uko is a news contributor for Tom's Hardware covering the latest updates in big tech and the PC industry. He is a mechanical engineer and senior technical writer with over nine years of experience in documentation and reporting. He is deeply passionate about all things engineering and technology, and is an expert in gadgets, manufacturing, robotics, automotive, and aerospace.

  • -Fran-
    The environmental impact of a, what will be in essence, "death beam" will surely be interesting to debate. Including Air Traffic around the area!

    Even if it'll be "low intensity", it'll still need a proper study around harms to the areas and risks. Also, low energy beams would require a lot of space so you make up for the low energy density.

    Interesting thought process, for sure.

    Regards.
    Reply
  • bill001g
    Meta has way too much money when they give it to companies that have some idea that has been around for years that nobody has figure out how to actually implement. They would have to have systems already being assembled to even think about having anything in orbit in 2 years. The time to get a slot from one the rocket launch companies is more than that. Almost seem like meta is being scammed. One of these companies meta is giving money to has not even existed for 4 years it is unlikely they have anything more than design ideas. It takes a very long time and lots of money to get even prototype boards from manufactures and they need stuff that can survive being launched into space.
    Reply
  • Gururu
    Doesn't that solar energy end up adding more heat to the earth whatever energy it is ultimately converted to?
    Reply
  • usertests
    Those satellites would harvest solar energy in space and beam it down to Earth-based solar farms using low-intensity near-infrared light.
    I remember China wanted to do it with microwaves: https://sustainabilitymag.com/articles/chinas-1km-solar-array-the-manhattan-project-of-energy
    Gururu said:
    Doesn't that solar energy end up adding more heat to the earth whatever energy it is ultimately converted to?
    It should be the same with nuclear fission/fusion, etc. If we started producing say 10,000x the energy that we use today, Earth wouldn't be able to radiate it away fast enough and we would cook.

    The global electricity production should be at least 20 TW (20,000 GW), so another 1 GW or 30 GW isn't going to have a significant effect.

    Space based solar doesn't produce greenhouse gases beyond construction and the rockets deploying the panels (maybe significant?). So it may be better than hydrocarbon sources.

    The real question: can it be cheap? Maybe if launch costs fall.
    Reply
  • edman565
    Wasn't this a type of power plant in one of the early Simcity games? Occasionally the beam would miss the collector and start fires so you had to keep open space around it.
    Reply
  • teckel12
    Isn't sunlight beamed down to Earth without any human involvement? Wouldn't the gain from solar collection in space be offset by the loss from beaming it through the atmosphere? It just seems that solar panels on Earth would be about as efficient once you factor everything. Not to mention the much higher cost and energy used to put solar collectors in space. This just doesn't seem financially or technically viable.
    Reply
  • edman565
    I think the idea is that the beam can work on the dark/night side of the planet if the solar reflector is at the right angle. This way it's a 24x7 supply.
    I am very skeptical that it would be worth it, but maybe the people paying for the experiment believe otherwise.
    Reply
  • rgd1101
    can't wait for the lawsuit
    Reply
  • SonoraTechnical
    just make it stop.
    i just hate every body remotely involved in tech now.

    Zuckerburg and the cohort of billionaires (Now we get to add Kevin O'Leary to the growing lists) are so anxious to convert our planet into Mordor.
    Reply
  • Notton
    The Soviets did the space mirror beam down to earth thing in the 90's.
    The prototype worked, but the followup failed to deploy.
    https://www.smithsonianmag.com/smart-news/how-russian-space-mirror-briefly-lit-night-180957894/
    While lithium-ion battery systems are effective for short-duration storage, they are generally not designed for sustained power delivery over multiple days.
    But where are they building a data center that you have to run on battery power for days?
    Reply