Close Menu
    Facebook LinkedIn YouTube WhatsApp X (Twitter) Pinterest
    Trending
    • Hisense U7SG TV Review (2026): Better Design, Great Value
    • Google is in talks with Marvell Technology to develop a memory processing unit that works alongside TPUs, and a new TPU for running AI models (Qianer Liu/The Information)
    • Premier League Soccer: Stream Man City vs. Arsenal From Anywhere Live
    • Dreaming in Cubes | Towards Data Science
    • Onda tiny house flips layout to fit three bedrooms and two bathrooms
    • Best Meta Glasses (2026): Ray-Ban, Oakley, AR
    • At the Beijing half-marathon, several humanoid robots beat human winners by 10+ minutes; a robot made by Honor beat the human world record held by Jacob Kiplimo (Reuters)
    • 1000xResist Studio’s Next Indie Game Asks: Can You Convince an AI It Isn’t Human?
    Facebook LinkedIn WhatsApp
    Times FeaturedTimes Featured
    Sunday, April 19
    • Home
    • Founders
    • Startups
    • Technology
    • Profiles
    • Entrepreneurs
    • Leaders
    • Students
    • VC Funds
    • More
      • AI
      • Robotics
      • Industries
      • Global
    Times FeaturedTimes Featured
    Home»Tech Innovation»Space data centers: AI’s next frontier explained
    Tech Innovation

    Space data centers: AI’s next frontier explained

    Editor Times FeaturedBy Editor Times FeaturedFebruary 7, 2026No Comments5 Mins Read
    Facebook Twitter Pinterest Telegram LinkedIn Tumblr WhatsApp Email
    Share
    Facebook Twitter LinkedIn Pinterest Telegram Email WhatsApp Copy Link


    Terrestrial information facilities are so 2025. We’re taking our large-scale compute infrastructure into orbit, child! Or not less than, that is what Large Tech is yelling from the rooftops in the intervening time. It is fairly a bonkers concept that’s hoovering up cash and mindspace, so let’s unpack what it is all about – and whether or not it is even grounded in actuality.

    Let’s begin with the fundamentals. You would possibly already know {that a} information middle is basically a big warehouse stuffed with hundreds of servers that run 24/7.

    AI corporations like Anthropic, OpenAI, and Google use information facilities in two principal methods:

    • Coaching AI fashions – That is extremely compute-intensive. Coaching a mannequin like those powering OpenAI’s ChatGPT or Anthropic’s Claude required operating calculations throughout hundreds of specialised chips (GPUs) concurrently for weeks or months.
    • Working AI companies – Whenever you converse with these fashions’ chatbots, your messages go to an information middle the place servers course of it and ship again the mannequin’s response. Multiply that by hundreds of thousands of customers having conversations concurrently, and also you want monumental computing energy prepared on-demand.

    AI corporations want information facilities as a result of they supply the coordinated energy of hundreds of machines working in tandem on these capabilities, plus the infrastructure to maintain them operating reliably across the clock.

    To that finish, these services are all the time on-line with ultra-fast web connections, and so they have huge cooling techniques to maintain these servers operating at peak efficiency ranges. All this requires lots of energy, which places a pressure on the grid and squeezes native assets.

    So what’s this noise about information facilities in house? The thought’s been bandied about for some time now as a vastly higher various that may harness infinitely considerable photo voltaic power and radiative cooling a whole bunch of miles above the bottom in low Earth orbit.

    Highly effective GPU-equipped servers can be contained in satellites, and so they’d transfer by way of house collectively in constellations, beaming information forwards and backwards as they journey across the Earth from pole to pole within the sun-synchronous orbit.

    The pondering behind house information facilities is that it will enable operators to scale up compute assets much more simply than on Earth. Up there, there are no constraints of simply out there energy, actual property, and contemporary water provides wanted for cooling.

    Starcloud has already begun operating and coaching a big language mannequin in house, so it might probably converse Shakespearean English

    There are a variety of corporations getting in on the motion, together with huge acquainted names and plucky upstarts. You have obtained Google partnering with Earth monitoring firm Planet on Project Suncatcher to launch a few prototype satellites by subsequent yr. Aetherflux, a startup that was initially all about beaming down solar power from space, now intends to make a data center node in orbit out there for industrial use early subsequent yr. Nvidia-backed Starcloud, which is targeted completely on space-based information facilities, despatched a GPU payload into house final November, and trained and ran a large language model on it.

    The newest to hitch the fold is SpaceX, which is set to merge with Elon Musk’s AI company xAI in a purported US$1.25-trillion deal with a view to usher within the period of orbital information facilities.

    In line with Musk’s calculations, it needs to be doable to extend the variety of rocket launches and the information middle satellites they will carry. “There’s a path to launching 1 TW/yr (1 terawatt of compute energy per yr) from Earth,” he famous in a memo, including that AI compute assets shall be cheaper to generate in house than on the bottom inside three years from now.

    As Elon Musk's SpaceX and xAI are set to merge, he's keen on increasing the number of satellite-carrying rocket launches per year to serve the need for space data centers
    As Elon Musk’s SpaceX and xAI are set to merge, he is eager on growing the variety of satellite-carrying rocket launches per yr to serve the necessity for house information facilities

    In a wonderful article in The Verge from final December, Elissa Welle laid out the numerous challenges these orbital data centers will have to overcome with the intention to function as marketed. For starters, they’d have to securely wade by way of the 6,600 tons of space debris floating around in orbit, in addition to the 14,000-plus lively satellites in orbit. Dodging these would require gas.

    You have additionally obtained to dissipate warmth from the space-based information facilities, and have astronauts keep them periodically. And that is to say nothing about how these satellites will have an effect on the work of astronomers or probably enhance gentle air pollution.

    In the end, there’s lots of experimentation and studying to be gleaned from these early efforts to construct out compute assets in house earlier than any firm or nationwide company can realistically scale them up.

    And whereas it’d finally change into doable to take action regardless of substantial difficulties, it is price asking ourselves whether or not AI is definitely on observe to profit humanity in all of the methods we have been promised, and whether or not we have to regularly construct out infrastructure for it – whether or not on the bottom or method up past the environment.





    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Editor Times Featured
    • Website

    Related Posts

    Onda tiny house flips layout to fit three bedrooms and two bathrooms

    April 19, 2026

    Efficient hybrid minivan delivers MPG

    April 19, 2026

    asexual fish defy extinction with gene repair

    April 19, 2026

    Rugged tablet boasts built-in projector and night vision

    April 19, 2026

    Powerful lightweight sports car available now

    April 19, 2026

    Adaptable medium format film camera changes sizes mid-roll

    April 18, 2026

    Comments are closed.

    Editors Picks

    Hisense U7SG TV Review (2026): Better Design, Great Value

    April 19, 2026

    Google is in talks with Marvell Technology to develop a memory processing unit that works alongside TPUs, and a new TPU for running AI models (Qianer Liu/The Information)

    April 19, 2026

    Premier League Soccer: Stream Man City vs. Arsenal From Anywhere Live

    April 19, 2026

    Dreaming in Cubes | Towards Data Science

    April 19, 2026
    Categories
    • Founders
    • Startups
    • Technology
    • Profiles
    • Entrepreneurs
    • Leaders
    • Students
    • VC Funds
    About Us
    About Us

    Welcome to Times Featured, an AI-driven entrepreneurship growth engine that is transforming the future of work, bridging the digital divide and encouraging younger community inclusion in the 4th Industrial Revolution, and nurturing new market leaders.

    Empowering the growth of profiles, leaders, entrepreneurs businesses, and startups on international landscape.

    Asia-Middle East-Europe-North America-Australia-Africa

    Facebook LinkedIn WhatsApp
    Featured Picks

    Legendary TVR Prepares Comeback With Griffith V8 Performance Car

    November 14, 2025

    OpenAI’s Sora Sparks Copyright Storm: Did AI Learn From Netflix and TikTok Without Permission?

    September 22, 2025

    OpenAI Rolls Back ChatGPT’s Model Router System for Most Users

    December 16, 2025
    Categories
    • Founders
    • Startups
    • Technology
    • Profiles
    • Entrepreneurs
    • Leaders
    • Students
    • VC Funds
    Copyright © 2024 Timesfeatured.com IP Limited. All Rights.
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us

    Type above and press Enter to search. Press Esc to cancel.