To The Moon

Finally, someone spoken up about the absurdity of an orbital data centre, without tackling all the pressing issues.

Naturally, it wasn't Musk, as he only speaks up when it benefits him.

Musk’s Orbital Data Center Statements

Below is the sequence of events where the snake oil salesman has spoken about it (In ascending date order)

• 31 Oct 2025 : Musk first publicly floated placing data centers in orbit in an X reply to an article discussing autonomous in-space construction.

• November 2025 : On “X”, Musk expanded on the concept, claiming orbital compute would become more economically viable than ground infrastructure within 3 to 5 years due to "unlimited solar power" and in-space radiative cooling.

• December 2025 : Following news leaks regarding a planned SpaceX listing, Musk confirmed pre-IPO discussions on social media, explicitly using "AI breakthroughs and space data center opportunities" to anchor pitch decks targeting a $1.5+ trillion valuation.

• Early 2026 : SpaceX submitted a formal authorization request to the FCC to deploy up to 1 million satellites equipped with orbital compute capabilities intended for edge processing and AI inference.

• May 2026 : The initial SpaceX prospectus listed space-based data centers as a long-term R&D initiative, noting a tentative timeline of "no earlier than 2028."

• June 8–10, 2026 : 2 days before SpaceX went public under the ticker SPCX on 12 Jun 2026, Musk aggressively pivoted the narrative.

• In press statements and X posts, he insisted orbital compute "is not a super hard problem" and promised to deploy 1 Gigawatt (GW) of space AI capacity by late 2027.

• Late August 2026 : Musk moved the target date forward again, announcing that the first batch of SpaceX "Starmind" AI satellites, packed with $NVIDIA(NVDA)$ chips, would make their maiden launch in Q4 2027.

• 01 Sep 2026: Via “X”, Musk publicly mocked technical critics, posting that self-described experts calling orbital AI "impossible due to cooling" fundamentally misunderstood thermal radiative formulas, coolant temperatures, and GPU operating thresholds.

Rebuttals & Technical Pushbacks.

The technical, financial, and logistical claims behind Musk's 2027 roadmap face sharp criticism across the sector:

(1) Sam Altman, CEO, OpenAI.

• Called orbital data centers “absurd” for now, pointing to launch costs versus terrestrial energy costs and the difficulty of repairing failing GPUs in orbit.

• He expects meaningful scale only later this decade.

(2) Taranis Reed, former NASA official.

• Argues that orbital data centers are “many years, perhaps decades, away” and requires launch costs below $200 per kg, about 7x lower than today.

• He also warns of collision & debris risks, as well as higher lifecycle carbon impact when accounting for build, launch & scrap.

(3) Jeff Bezos, $Amazon.com(AMZN)$.

• Shared at VivaTech Conference in Paris on 17 Jun 2026 that he believes orbital compute will eventually become viable in the long run.

• Citing costs as the single biggest barrier, in addition to (space) launch economics and satellite manufacturing costs drop drastically.

• Until then, ground-based data centers remain the rational choice, noting (optimistically) that "eventually the lines will cross" but current launch expenses and specialized hardware costs mean the economics do not make sense today.

(4) Masayoshi Son, $Softbank Corp.(SOBKY)$.

• He has said space data centers will not cut costs soon enough to matter in the AI race, where the next few years are critical.

(5) According to global consultancy firm - Wood Mackenzie:

• Engineering constraints including (a) the need for radiative heat rejection in vacuum, (b) radiation tolerance to cosmic rays & solar wind, (c) the challenge of moving massive AI datasets to and from orbit, and (d) the sheer capital intensity.

• Costs estimates to operate a a 1.0 GW space data center will costs around $170 billion, that is more than 3x the cost of a earth-based equivalent, do not make sense.

(6) Blaine Curcio of Orbital Gateway Consulting and Evely Chow of Neuberger Berman - have leaned in on 4 main obstacles enroute to a sustainable space data centre.

• Cooling in vacuum: Unlike ground-based facilities that use liquid cooling, space-based data centers must dissipate heat without air or water and also withstand radiation.

• Rapid obsolescence: GPUs and AI hardware are evolving so fast that a state‑of‑the‑art orbital data center could be outdated within a few years of launch, despite enormous upfront costs.

• High‑volume data downlink: Building the data center is only half the problem; getting massive volumes of data back to Earth reliably is critical. Companies like Transcelestial are “working” on laser-based links to solve this.

• Powering hyperscale: Reaching true hyperscale capacity in orbit may require nuclear power and is seen as 5 - 7 years away at best.

Storytelling again ?

Is $SpaceX(SPCX)$’s orbital data centre a case of déjà vu ?

There are structural parallels to his TSLA’s autonomous‑driving narrative and it warrants skepticism:

Timeline slippage and moving targets:

• SPCX’s May 2026 S‑1 registration form listed deployment “as early as 2028.”

• By late August 2026, Musk publicly advances to demo launches in Q4 2027 and “significant scale” in 2028.

• The compression mirrors Musk’s prior patterns where ambitious dates are announced before capital events (IPO, earnings) and then walk back or slide.

Regulatory & technical headwinds.

• SPCX’s own IPO filing flags reliance on “unproven technologies, or technologies that do not exist” for orbital data centers.

• Independent engineers emphasize unsolved problems in cooling, radiation hardening, serviceability, and cost - issues that don’t disappear with rhetoric.

Economic logic vs Storytelling incentives.

• Multiple analysts say meaningful scale is a 2030s outcome, not late‑2027.

• Yet the narrative is front‑and‑center in SPCX’s IPO window and subsequent equity promotion, where valuation benefits from AI infrastructure optionality.

Case study - TSLA’s Autonomous driving:

• Despite years of promises, TSLA’s full self‑driving (FSD) remains constrained by regulation and technical limits; the gap between demonstration and deployable, regulator‑approved autonomy has proven large.

• Space data centers face even harsher physics (radiation, thermal rejection, no in‑orbit repair) and higher capital intensity, making the “2 to 3 years to cost parity” claim especially aggressive.

Overall, looking at (1) Musk's bold promises, (2) the cautious warnings in official filings, and (3) experts’ consensus of a later timeline - it is clear that Musk’s space data center schedule is at best, highly optimistic or at worst, a stunt to keep investors hyped about SPCX and its AI growth.

Led by the Nose, again?

Do investors genuinely believe Musk will deliver a commercially viable space data center by 2027 ?

Or are they simply willing to be led by their noses once more, trading one unfulfilled dream for another ?

Autonomous driving is a software challenge constrained primarily by earthly environment variables—a problem far simpler than launching gigawatts of overheating, radiation-vulnerable silicon into vacuum orbits.

Yet, after a decade of missed deadlines, regulatory crackdowns, and unkept robotaxi promises on Earth, market participants eagerly handed SPCX a record-breaking $2 trillion market valuation based on an even grander cosmic fantasy.

If terrestrial autonomy—where roads, signs, and atmosphere are readily available—remains an elusive target after a decade of grandstanding, expecting space data centers to scale by 2027 isn't an investment strategy; it is a suspension of disbelief.

Question remains: how many times will Wall Street buy the narrative before demanding the laws of physics and execution be met?

SPCX a technical ‘Buy’ ?

With reasonable doubt that SPCX’s touted orbital data centre will likely not happen by 2027, is SPCX still a ‘Buy’ ?

To answer this question stripped of emotions, a look at SPCX’s technical indicators of (a) Simple moving averages, (b) MACD and (c) RSI hopefully will shed some light.

(1) Simple Moving Average (SMA).

On Tue, 08 Sep 2026 intraday trading, SPCX was at $153.03 per share - higher than its 3 SMAs of (a) 10-day ($141.95), (b) 30-day ($132.60) and (c) 50-day ($135.98), above its SMAs.

Additionally, the shorter-term 10-day SMA has crossed above both the 30-day and 50-day SMAs.

This bullish alignment suggests that short-term buying momentum is outpacing medium-term trends, establishing a solid baseline of dynamic support below current price levels.

(2) MACD.

The MACD line (2.74) and Signal line (1.33) are positioned above the center Zero line, confirming that the overall medium-term trend is in positive territory.

Additionally, with the MACD line tradings above the Signal line - this positive relationship reflects active, short-term bullish momentum leading the underlying average trend.

Conclusion.

The positive divergence reading of 1.41 signifies expanding positive momentum, with the widening gap confirms that buying volume is driving the trajectory upward toward recent highs rather than slowing down.

After all that has been said, I think SPCX can still merit investor interest.

However, its appeal should rest on execution (today) and not a space data-centre promise that may not happen, just like TSLA’s autonomous driving.

Latest technical strength may reflect improving sentiment, yet only tangible commercial progress can justify a lasting re-rating. Agree ?

Remember to check out my other posts. (See below). Help to Repost ok, Thanks.

Must Read: Click on below titles to access. Repost to share, Like as encouragement ok. Thanks.

• TSLA, cont'd to fall this Week or Rally ?

• US market drags down by mixed US reports !

• DELL, the next Magnificent 7 ? Really ?

• Do you think still believe that Musk is able to deliver its autonomous robotaxi commercially ?

• Do you think still believe that Musk can deliver a space data centre by 2027 ?

If you find this post interesting, give it wings! ️ Repost and share the insights ?

Do consider “Follow me” and get firsthand read of my daily new post. Thank you.

@Daily_Discussion

@TigerPM

@TigerStars

@Tiger_SG

@TigerEvents