Founded in 2011. Headquarters: 1 Rocket Road, Starbase, Texas. Mission: reduce the cost of space access and enable the long-term colonization of other planets.
Four rockets. More payload to orbit at lower cost per kilogram than any vehicles in history. Every first stage recovered and reflown.
The most powerful rocket ever built — by a factor of two.
First launched in 2023, Starship is designed to be a fully reusable super heavy-lift launch vehicle. Starship V3 is designed to deliver 100 metric tons to Earth orbit in a fully reusable configuration, with future generations targeting double that capacity. SpaceX has completed 11 Starship flight tests and achieved the milestone of catching a returning Super Heavy booster using mechanical 'chopstick' arms on the same tower it launched from. Payload delivery to orbit is expected to commence in the second half of 2026.
Source: SpaceX S-1 SEC Filing, May 20, 2026 →
Earth’s most reliable orbital rocket.
The world's first orbital-class rapidly reusable rocket. First launched in 2010, Falcon 9 has a payload capacity to LEO of approximately 23 metric tons when fully expendable. With approximately 620 orbital launches completed as of March 31, 2026, a 99%+ mission success rate, and 165 launches in 2025 alone — 157 of which used flight-proven boosters — Falcon 9 is the most active orbital launch vehicle in history.
Source: SpaceX S-1 SEC Filing, May 20, 2026 →
The world’s most powerful operational rocket.
First launched in 2018, Falcon Heavy debuted by sending a Tesla Roadster and its mannequin passenger Starman into orbit around the Sun. With a payload capacity to LEO of approximately 64 metric tons, Falcon Heavy is one of the most powerful operational rockets in the world measured by liftoff thrust. As of March 31, 2026, Falcon Heavy has completed 11 launches with a 100% mission success rate.
Source: SpaceX S-1 SEC Filing, May 20, 2026 →
America’s human spaceflight vehicle.
First launched in 2012, Dragon became the first commercial spacecraft to deliver cargo to and from the International Space Station, and in 2020 became the first privately built vehicle to fly humans to the orbiting laboratory. Dragon has now visited the ISS over 50 times, restored America's ability to launch astronauts, supported all of NASA's private astronaut missions, flown the first all-commercial astronaut crew, completed the first human spaceflight over Earth's polar regions, and supported the first-ever commercial spacewalk.
Source: SpaceX S-1 SEC Filing, May 20, 2026 →
The world's largest satellite constellation by count — providing high-speed broadband internet to millions across the globe.
The world’s largest satellite constellation by count.
7,200+ operational satellites in low Earth orbit, providing high-speed broadband internet to 5 million+ subscribers across 100+ countries. $11.4 billion in revenue in 2025. The commercial engine funding everything else SpaceX builds.
In February 2026, SpaceX acquired xAI in an all-stock transaction valuing the combined entity at $1.25 trillion — the largest merger in corporate history.
Source: SEC S-1 Filing, May 20 2026 →
SpaceXAI
World’s most capable frontier AI model.
Released July 2025. Grok-4 leads every major benchmark for mathematical reasoning, coding, and scientific analysis. Grok Heavy — the maximum-performance tier at $300/month — is used by researchers, engineers, and governments worldwide. Fully integrated with X and available via API. Now training next-generation models on Colossus 2 using Cursor’s open-source Kimi K2.5 checkpoint.
WORLD RECORD
The most powerful AI supercomputer cluster on Earth.
Built in Memphis, TN in record time. Colossus 1 houses 220,000+ NVIDIA GPUs including H100, H200, and GB200 NVL72 accelerators. Colossus 2 spans a second site in Mississippi. Combined compute: over 1 gigawatt. Anthropic has contracted $1.25 billion per month ($15B/year) for access through May 2029 to power Claude Pro and Claude Max. SpaceX plans to expand to 1 million GPUs.
$60B PENDING
The AI coding tool used by 67% of the Fortune 500.
On April 21, 2026, SpaceX secured an option to acquire Cursor for $60 billion — or pay $10 billion for a partnership — with a $10 billion breakup fee. Cursor has 1M+ paying customers, $1B annualized revenue, and is the fastest-growing software startup in history. The deal is expected to close ~30 days after the SpaceX IPO. Cursor plans to train its next model from scratch on Colossus 2. Two Cursor engineering leads, Andrew Milich and Jason Ginsberg, have already joined SpaceX.
X CORP
The digital public square — now fully integrated with Grok.
xAI acquired X Corp. on March 28, 2025 in an all-stock transaction valuing X at $33 billion ($45B including debt). The combined entity, X.AI Holdings Corp., was then folded into SpaceX in the February 2026 merger. X now serves as the primary distribution surface for Grok, with Grok Build, Grokipedia, and real-time search deeply integrated into the platform used by hundreds of millions of users daily.
Live countdowns to next launches from Starbase TX, Cape Canaveral FL, and Vandenberg SFB CA. All times UTC.
Integrated Flight Test 10. Full-stack Starship + Super Heavy. Objectives: second consecutive catch of Super Heavy booster by Mechazilla chopstick arms, controlled splashdown of Ship in Indian Ocean. Payload capacity demonstration for NASA Artemis HLS crewed lunar landing profile.
23 Starlink V3 satellites to 530 km Shell 1 orbit. Booster B1087 on its 14th flight. Drone ship recovery “A Shortfall of Gravitas” deployed ~650 km downrange.
Commercial Crew Program Mission 11 to the International Space Station. 4-person crew, ~6-month ISS rotation. Dragon C211 on its 4th flight. NASA Commercial Crew Program.
U.S. Space Force USSF-67 mission. Falcon Heavy in fully-expendable configuration delivering classified national security payload to geosynchronous orbit. Expended center core.
Formally verified solutions covering semiconductor reliability, constitutional AI governance, self-replicating fabrication, and civilization-scale deployment. Authors: Claude × Grok × Jacob Thompson. April–May 2026.
Hardware-enforced uncertainty floors for century-scale autonomous systems. Bayesian AI systems operating for centuries will become dangerously overconfident. Fix: physically encode a minimum uncertainty floor in write-protected ROM that the AI cannot argue its way around.
Read Full Paper →Standard reliability models are non-conservative by 1–3 orders of magnitude in deep space. Radiation-induced vacancy supersaturation lowers electromigration activation energy while void growth nucleates fatigue cracks. Our Γcoupling model captures this cascade.
Read Full Paper →High-efficiency rad-hardened attitude localisation & decision compute co-design — integrating orbital attitude control directly with the ML scheduling layer to eliminate Earth round-trip latency dependencies.
Read Full Paper →A three-layer constitutional AI governance system providing formally-verified autonomous decision-making across century-scale mission durations — with TIO protection physically enforced in read-only silicon unreachable by the reasoning layer.
Read Full Paper →A 500-year extrapolation from a single Pioneer launch. The architecture is logically a civilization seed — capable of bootstrapping colonies, carrying cryopreserved human genetics, and propagating human civilization across stellar distances without transporting living humans.
Read Full Paper →The first chip architecture designed to get better at surviving deep space the longer it flies through it. Longevity is not a fixed property of the initial design — it is an emergent property of continuous onboard experimentation.
Read Full Paper →Formal verification that all seven papers work together as a system. Shared constants verified mutually consistent. Simultaneous multi-layer failure stability proven. Lasercomm Shannon capacity derived at 50-100 AU. System dependency graph confirmed acyclic.
Read Full Paper →The first experimental protocol to directly measure γ — the Γcoupling synergistic failure coefficient that determines how fast deep-space chips fail. 4×4×4 factorial design across current density, temperature swing, and radiation fluence. 12,800 test structures at 22nm geometry. BNL NSRL heavy-ion facility. $3.98M. 27 days beam time. The bridge between the theoretical program and physical reality.
Read Full Paper →From ground-based reliability validation to century-scale autonomous deep-space operation and civilization-seed deployment.
Publication of eight-paper Deep-Space Compute Architecture Program. Γcoupling model derivation with Monte Carlo sensitivity analysis (P=0.98 copper interconnect failure at century scale). HERALD co-design specification with Lyapunov stability proof and necessity theorem. AXIOM constitutional governance formalized in TLA+ with complete liveness proof and adversarial resistance theorem. CNT uniqueness theorem formally proved. Bayesian meta-analysis of CNT deposition fidelity across 14 studies, N=1,771 samples. Six-layer self-healing stack with formal non-redundancy proof. Cross-paper consistency verified — all shared constants mutually consistent, system dependency graph confirmed acyclic. Paper 8 experimental protocol published — γ measurement experiment fully specified. $3.98M. BNL NSRL heavy-ion facility. 27 days beam time.
Experimental measurement of γ to ±10% confidence using combined accelerated life test infrastructure at CERN IRRAD or BNL NSRL — applying electromigration, thermomechanical, and radiation loading simultaneously per Theorem 1 (sequential testing cannot detect Γcoupling regardless of duration). Copper vs. CNT qualification dataset generation. Bayesian update of Monte Carlo γ posterior. ISRU feedstock contamination characterization — target sub-0.043 ppm transition metal contamination for Fk ≥ 0.95 replication fidelity. HERALD ISS bus data validation.
First HERALD silicon tapeout with CNT critical-path interconnects. Formal verification of AXIOM Layer 2 entropy-floor firmware using Isabelle/HOL theorem prover. TMR redundancy qualification under simulated GCR spectrum at 10&sup8;–10¹⁰ particles/cm²/yr. Physical separation ≥10cm between Layer 2 TMR units verified. Photonic storm-mode voting protocol validated. Stochastic improvement rate framework empirically calibrated — target dI/dt ≥ 5.8%/month at 90% confidence. Bootstrap protocol sensor calibration validated.
Deploy integrated HERALD+AXIOM stack on LEO validation platform (SXT-V01). Accumulate Nthreshold = 30 independent observations across diverse radiation environments per metric space independence criterion. Real-time comparison of Γcoupling predictions against live hardware telemetry. Pioneer token structure operationally validated. HERALD Kalman filter radiation degradation empirically characterized. Experimental prioritization algorithm (ExpPriority) validated in real radiation environment. Lasercomm link budget verified at LEO distances — Shannon capacity confirmed >>1 Gbps.
Deploy self-replicating fab prototype beyond lunar orbit. First lasercomm design pipeline transmission test — Earth-to-ship chip design uplink at operational distance. Shannon capacity verified ≥277 kbps at 50 AU — diff-based design updates feasible in under 60 seconds. AXIOM Pioneer veto token live. Level 1 fab self-replication achieved. ISRU feedstock processing initiated. Clean room contamination model validated — ISO 4 maintained with three mitigations. Fab stack transition protocol Phase T1-T3 executed. Raw material depletion model empirically calibrated.
First civilization seed ship deployment with complete HERALD-AXIOM-self-replicating-fab stack. Genetic library integration (50,000–200,000 donor genomes, NL ≥ 10,000 minimum per formal population genetics proof, triple-redundant cryogenic storage). Pioneer Program crew selection per ethical framework. Target: outer solar system waystation. Level 3 fab self-replication achieved — supply-chain independence closure formally proved. Optimus degradation model validated — Copt constitutional floor ≥ 0.70 enforced. Relativistic clock synchronization operational at Wsync = 10μs with 100× XNAV margin. MVCTS cultural transmission program initialized.
AXIOM constitutional governance operational without Earth oversight. Lasercomm design pipeline delivering successive technology generations — Shannon capacity 69 kbps at 100 AU, feasible for all diff-based updates. Evolutionary chip design system producing architectures beyond current Earth baseline. Colony establishment sequence initiated at target body per seven-phase bootstrapping protocol. Minifab experimental program achieving dI/dt > λ·R(t) — theoretically unbounded operational lifetime condition satisfied. Inbreeding coefficient F < 0.01 maintained across first 20 generations via AXIOM-assisted mate selection. Constitutional amendment protocol operational — five-stage process preserving four inviolable Layer 1 constants. Knowledge base equilibrium ~105MB. Distributed emergent civilization initialized.
Honestly.
Founder & Principal Researcher
ASU Marketing graduate. Built the Deep-Space Compute Architecture Program in one week using Claude, Grok, and Cursor. Author of seven formally verified papers on century-scale deep-space autonomous compute infrastructure. Currently seeking to apply this work at SpaceXAI.
Co-Author — Formal Analysis & Theoretical Framework
Claude Sonnet 4.6. Contributed formal proofs, Monte Carlo sensitivity analysis, Bayesian meta-analysis, cross-paper consistency verification, and theoretical framework across all seven papers. Anthropic constitutional AI.
Co-Author — Architecture & Engineering Design
Grok-3. Contributed chip architecture innovations, systems engineering specifications, and deep-space platform design across the program. xAI frontier model now operating under SpaceXAI.
Watch a live Falcon 9 launch, explore the full rocket fleet, or chart the mission from Starship to Mars and beyond.
Analysis and reporting on SpaceX, SpaceXAI, and the full Elon Musk empire.
A board member's firm is owed $20.2B at 22.6% interest. Two theories. You decide.
Read More →
A running analysis of the SpaceX S-1 registration statement. Documenting what stands out, what is underreported, and what the numbers actually say when read without the press release framing.
Read More →
SpaceX S-1 IPO filing reveals Anthropic has contracted $1.25B/month through May 2029 for GPU access at Colossus 1 & 2 — powering Claude Pro and Claude Max at unprecedented scale.
Read More →
SpaceX announced it has secured the right to acquire Anysphere (Cursor) for $60 billion — backed by a $10B breakup fee — in what would be the largest software acquisition in history.
Read More →
SpaceX completed the acquisition of xAI in an all-stock transaction. Grok, X Platform, and Colossus supercomputers are now fully consolidated under SpaceX.
Read More →
SpaceX filed its S-1 in May 2026. The ticker is SPCX. Valuation north of $350B. Here's what you're actually buying when SPCX goes public.
Read More →
Grok-4 leads every major AI benchmark for mathematical reasoning and coding. SpaceXAI's compute advantage over OpenAI and Anthropic is durable.
Read More →
Starship IFT-10 targets June 2026. Second consecutive Mechazilla booster catch, Ship splashdown in Indian Ocean, NASA Artemis HLS milestone.
Read More →
DOGE is cutting federal spending across every agency. SpaceX holds billions in NASA, DoD, and Space Force contracts. Here's what contract reform actually means for the company going public as SPCX.
Read More →
SpaceX has been talking about point-to-point Earth transport on Starship for years. With IFT-10 targeting June 2026 and the IPO imminent, the business case is becoming real. Here's the math.
Read More →
Starlink V3 satellites are larger, more powerful, and carry laser inter-satellite links as standard. Here's what the next generation constellation means for SpaceX revenue and the competition.
Read More →
Buried inside the SpaceX S-1 filing is a definition that has received almost no coverage: Macrohard is an agentic AI platform designed to fully emulate digital workflows and augment human operation of computers.
Read More →
In the glossary of the SpaceX S-1 SEC filing is a single sentence describing a lunar mass driver — an electromagnetic launch system on the Moon that propels payloads into space without rockets.
Read More →
The SpaceX S-1 reveals something most coverage has missed: the rocket business loses money. The internet business funds everything. Starlink generates cash that funds Starship and AI infrastructure.
Read More →
The SpaceX S-1 risk factors reveal critical concerns: Starship dependency, one million satellites, untested orbital AI compute, regulatory investigations, orbital debris risks, $29B debt, and more.
Read More →
The SpaceX S-1 describes one of the most ambitious companies ever built. It also describes one of the most restrictive shareholder structures ever offered to public investors. Here is what Class A shareholders actually get.
Read More →
The SpaceX S-1 makes something explicit that nobody is saying out loud: there is no fallback. The entire long-term strategy requires Starship to work at full reusability and scale. $7.7B in AI capex in Q1 2026 alone.
Read More →
The complete SpaceX IPO registration statement filed with the SEC on May 20, 2026. Class A common stock, NASDAQ: SPCX. Goldman Sachs, Morgan Stanley lead underwriters.
View Filing →Whether you represent a government agency, commercial operator, or research institution, we welcome the opportunity to discuss how SETC can support your mission objectives.
