The Robinhood of Space Is Going for Broke

Barely 18 months after its public debut, Aetherflux — the space-based solar power startup founded by Robinhood co-founder Baiju Bhatt — is in talks to raise $250 million to $350 million in a Series B round that would value the company at roughly $2 billion. Index Ventures is reportedly leading the deal, following up on its co-lead position in the company's $50 million Series A just under a year ago.

The valuation represents a staggering ~30x markup from the company's total funding to date of approximately $60 million. But the number starts to make more sense when you understand what has changed: Aetherflux is no longer just a space solar company. It's positioning itself as the energy backbone for AI computing in orbit — and it has NVIDIA's backing to prove it.

From Wall Street to the Final Frontier

Baiju Bhatt's path to space entrepreneurship has been anything but conventional. The son of a NASA Langley Research Center scientist, Bhatt grew up steeped in aerospace before pursuing degrees in physics (B.S.) and mathematics (M.S.) at Stanford University. It was at Stanford that he met Vlad Tenev, and the pair went on to co-found Robinhood in 2013, democratizing stock trading and building a company that would reach a peak valuation of over $40 billion.

Bhatt served as Robinhood's co-CEO until November 2020, then transitioned to Chief Creative Officer before departing the executive team entirely in March 2024. Seven months later, he unveiled Aetherflux — a return to his first love.

"This is something I've been passionate about since childhood," Bhatt has said, pointing to his father's NASA career as the original spark. He put roughly $10 million of his own money into the company at inception, a move that helped attract an elite roster of investors.

The Technology: Harvesting Sunlight in Orbit

Aetherflux's core technology is space-based solar power (SBSP) — a concept that has existed since the 1960s but has never been commercially realized. The physics are compelling: solar energy in space is approximately 8x more intense than on Earth's surface, unfiltered by atmosphere and available nearly 24/7.

The company's approach differs from legacy SBSP proposals in two critical ways:

• Low Earth Orbit (LEO) constellation rather than geostationary orbit, enabling faster deployment and lower per-satellite costs
• Infrared laser transmission instead of microwaves, allowing compact ground receivers of just 5-10 meters that can be deployed in hours

The plan calls for launching roughly 30 satellites per SpaceX Falcon 9 rideshare mission, eventually scaling to thousands of satellites forming a "power grid in space." Initial target customers include the U.S. military (which lost over 3,000 personnel to fuel convoy attacks in Iraq and Afghanistan between 2003-2007), remote research stations, and mining operations. The Department of Defense's Operational Energy Capability Improvement Fund has already approved funding for a proof-of-concept demonstration.

The Pivot That Changed Everything: Galactic Brain

In December 2025, Aetherflux announced Galactic Brain — an orbital data center project that would use the company's space-based solar infrastructure to power AI computing hardware directly in orbit. The first data center satellite is targeted for commercial operation by Q1 2027.

The strategic logic is elegant: rather than solving the enormous challenge of beaming power from space to Earth (something humanity has never achieved at commercial scale), why not bring the compute to the power source?

The timing couldn't be better. Global data centers consumed 415 terawatt-hours in 2024, and Goldman Sachs projects a 160% increase in data center power demand by 2030. Every major hyperscaler is scrambling for energy — Microsoft is restarting Three Mile Island, Amazon is buying nuclear, and Google is investing in geothermal. Aetherflux is proposing a radically different solution: skip the terrestrial grid entirely.

The pivot received a massive credibility boost at NVIDIA's GTC 2026 conference in March, where Jensen Huang unveiled the company's space computing initiative. NVIDIA announced the Space-1 Vera Rubin Module, delivering 25x more AI compute than the H100 GPU for orbital inference tasks. Aetherflux was named one of just six launch partners, alongside Axiom Space, Planet Labs, and Kepler Communications.

"NVIDIA Space-1 Vera Rubin Module delivers high-performance, energy-efficient AI at the edge in orbit, powered by solar energy," Bhatt said at the event.

The Series B: Following the Smart Money

Aetherflux's fundraising trajectory tells a story of rapidly escalating conviction from top-tier investors.

Series A (April 2025) — $50 million:

• Co-leads: Index Ventures, Interlagos (VC founded by former SpaceX executives)
• Participants: Breakthrough Energy Ventures (Bill Gates), Andreessen Horowitz, New Enterprise Associates
• Notable individuals: Vlad Tenev, Daniel Gallagher (former Robinhood CLO), Jared Leto

Series B (in discussion) — $250-350 million at $2B valuation:

• Lead: Index Ventures
• Other participants: Not yet disclosed

The fact that Index Ventures is leading both rounds is a strong signal. In venture capital, when a lead investor doubles down at a dramatically higher valuation, it typically indicates deep due diligence and high conviction. The involvement of Interlagos — founded by former SpaceX leadership with intimate knowledge of orbital economics — adds another layer of credibility.

Market Context: Riding Two Mega-Trends

Aetherflux sits at the intersection of two rapidly expanding markets:

Space-Based Solar Power: Valued at $3.46 billion in 2025, projected to reach $10.70 billion by 2035 (11.95% CAGR), according to SNS Insider. A recent UK government study found that SBSP could be cost-competitive with terrestrial solar by 2040, with levelized costs potentially dropping to £0.0087-0.0129/kWh.

Orbital Data Center Infrastructure: Projected to grow from $1.77 billion in 2029 to $39.1 billion by 2035 — a staggering 67.4% CAGR. This nascent market is being driven by the insatiable demand for AI compute and the growing energy constraints facing terrestrial facilities.

The competitive landscape is intensifying. Overview Energy raised $20 million in December 2025 for a geostationary approach. Mantis Space closed a $10 million seed round in March 2026 for laser-based satellites in Medium Earth Orbit. The UK's Space Solar has tested a 1.8km modular solar array. Japan's JAXA is planning a small-scale demonstration in FY2026, and China is pursuing both a kilometer-wide orbital solar station and a 2,800-satellite compute constellation.

But Aetherflux has a structural advantage that competitors lack: dual revenue streams from a single technology stack. The same satellite infrastructure can sell power to Earth and host orbital AI compute, creating optionality that pure-play competitors cannot match.

The Bull and Bear Cases

The bull case is straightforward: AI's energy problem is real, urgent, and getting worse. If Aetherflux can demonstrate commercially viable orbital computing by 2027, it could become critical infrastructure for the AI era — a "picks and shovels" play on the biggest technology trend of the decade. The NVIDIA partnership provides both technical validation and a potential distribution channel.

The bear case centers on execution risk. Beaming energy from space at commercial scale is something humanity has literally never done. Orbital data centers face extreme engineering challenges — thermal management, radiation hardening, latency, and maintenance are all unsolved at scale. The $2 billion valuation prices in significant success that remains to be demonstrated. And competition from deep-pocketed players like SpaceX and Amazon (both exploring orbital compute) could erode any first-mover advantage.

There's also the question of whether the pivot itself signals that the original earth-transmission business model was harder than anticipated. Converting laser power beamed from orbit into usable electricity at commercial scale remains an unproven technology.

What to Watch

Aetherflux's story will be defined by three near-term milestones: the 2026 orbital demonstration mission (proving end-to-end power transmission from space), the Q1 2027 launch of the first data center satellite (proving orbital compute viability), and the closing terms of the Series B (confirming market conviction at the $2 billion valuation). If the company hits all three, Bhatt will have achieved something arguably more audacious than Robinhood — building the energy infrastructure for an AI economy that extends beyond Earth. If it stumbles, the $2 billion price tag will look like peak hype. Either way, the space solar sector has never had this much attention, capital, or urgency. That alone makes Aetherflux one of the most important startups to watch in 2026.

---

Sources:

• TechCrunch — Aetherflux reportedly raising Series B at $2 billion valuation
• Aetherflux — Wikipedia
• NVIDIA Newsroom — Space Computing
• SpaceNews — Aetherflux enters orbital data center race
• EntrepreneurLoop — Aetherflux Eyes $2 Billion Valuation
• SNS Insider — Space-Based Solar Power Market Size
• Payload Space — Orbital Data and Niche Markets Give Space Solar a New Shimmer