SPACEX is undervalued if it follows the revenue and profit surge of Micron, it will be quickly 10X share price over 12-24 months.
SPACEX is 4 Months away from boostings its AI revenue from $1 billion per year in April to $33 billion per year runrate in October 2026.
Gigawatts times Billions per GW = Revenue per GW
They have a 1.1 gigawatts of AI data center as of now. They are renting 800MW for $26 billion per year.
They will have 1.5 gigawatts around october 2026. The S1A indicating they had not actually installed all 550000 chips as of April 2026.
0.3 GW at $20 billion per GW is $7 billion per year (Hopper chips)
0.4 GW at $50 billion per GW is $20 billion per year (Blackwell Chips)
AI revenue from Cursor and Grok make up the rest.
0.2 GW is still being used for Cursor and Grok until October.
0.4 GW will be added with 220K B300 before Google starts renting all of the capacity.
0.7 GW at $50 billion per GW for B300 remaining in Colossus 2. This is coming online in about 3-4 months.
At COLOSSUS II, they brought online the first cluster of approximately 110,000 GB200 processors, approximately 210 megawatts of compute power, even faster in 91 days. As an illustrative comparison, an industry benchmark to bring online a 100 megawatt greenfield data center is approximately two years. Furthermore, in the case of COLOSSUS II, following the initial cluster, they brought online the second cluster of 110,000 GB300 processors and 220 megawatts of compute power in 64 days, demonstrating the ability to rapidly scale facilities once built. They expect that once fully operational, the next phase of expansion at COLOSSUS II will bring online at least 220,000 additional GB300 processors and over 400 additional megawatts of compute power.
0.4 GW at $50 billion per GW could be added every 3 months or so.
This is before Multipliers Per Energy and Multipliers of revenue per GW of energy.
2027 Rubin 3X rent per GW, 10X inference tokens
Rubin will be $100-$120 billion per year for rental.
Rubin chips will be about $1T per year per GW if you have frontier or top model.
2026-27 AI Catchup (Cursor) 6X (frontier is more profit than renting) @elonmusk
If Cursor and Grok become frontier models.
1 GW would generate $300 billion per year using Blackwell.
1 GW would generate $1 trillion per year using Rubin if it is frontier
GW Multipliers – way more energy
SPACE AI 1.3X 2027, 2X 2028, 10X 2029
4GW Q42027, 1 GW Space
2+GW per year each year 2027, 2028, 2029
2028 4-10 GW Space
SpaceXAI has 6+GW of new data center power locked up for the next 3 years. Will have 2 GW by the end of this year.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
