来自 Maksym Misichenko · Nasdaq ·
AI智能体对这条新闻的看法
The panel consensus is bearish, with concerns about SpaceX's high valuation, widening losses, and slowing growth. While there's debate on the bundling effect of Starlink and xAI, the panel agrees that the current IPO target of $1.8T may not be justified.
风险: The widening R&D losses from the xAI integration and the slowing growth of Starlink's ARPU.
机会: The potential bundling effect of Starlink's satellite constellation becoming xAI's captive, low-latency compute backbone.
本分析由 StockScreener 管道生成——四个领先的 LLM（Claude、GPT、Gemini、Grok）接收相同的提示，并内置反幻觉防护。 阅读方法论 →
據報導，SpaceX 將其 IPO 估值目標從 2 萬億美元下調至 1.8 萬億美元。
其 S-1 報告顯示，其增長正在放緩，虧損正在擴大。
根據傳統指標，該公司看起來價格過高。
就在 SpaceX 發布其 S-1 報告幾天後，該公司已經開始縮小其 IPO 目標。
Elon Musk 的太空探索、衛星互聯網和人工智能公司現在據報導正在以至少 1.8 萬億美元的估值進行公開發行，此前目標是至少 2 萬億美元。
人工智能會創造世界上第一個萬億富翁嗎？ 我們的團隊剛剛發布了一份關於名為“不可或缺的壟斷”的、為 Nvidia 和 Intel 都需要的關鍵技術的報告。繼續 »
在 X 上的一篇帖子中，Musk 表示這個消息是“虛假的”，目前尚不清楚降低目標是否是對投資者興趣不足的回應，儘管它可能對這個太空巨頭來說是一種必要的現實檢驗。
圖片來源：SpaceX。
SpaceX 的 S-1，所有計劃公開上市的公司都必須提交的 SEC 檔案，主要是一場失望。它顯示出一家增長緩慢且虧損擴大的公司，尤其是在收購 xAI 之後。
2025 年，SpaceX 的收入增長了 33% 至 187 億美元，但報告了符合一般公認會計原則 (GAAP) 的營業虧損 26 億美元，與 2024 年的 4.66 億美元利潤相比，這是由於研發支出激增，原因是其人工智能部門（前身為 xAI）的支出增加。
在 2026 年第一季度，收入增長放緩至 15.4% 至 47 億美元，其 GAAP 營業虧損為 19.4 億美元，再次由於研發費用增加一倍多至 35 億美元。
然而，SpaceX 認為其估值由其未來決定，包括殖民火星和月球的任務，而不是其目前的業績。
該公司聲稱擁有 28.5 萬億美元的可執行總潛在市場規模，大致相當於美國 GDP，並稱其為人類歷史上最大的潛在市場。其中大部分來自人工智能企業應用，該應用價值 22.7 萬億美元，是根據數字合作組織估計的全球數字經濟規模。
如果您希望在 SpaceX IPO 上市時買入，那麼更低的价格是好事，但公司可能在華爾街高估其價值的一個早期信號令人擔憂。
在市盈率上，SpaceX 的交易價格接近 100，比 S&P 500 上的每支股票都貴。它還在虧損。
雖然業務中有一些光明面，例如其 Colossus I 和 II 數據中心對計算的需求，SpaceX 在與 xAI 合併中獲得了這些數據中心，但也有一些問題的跡象。
例如，xAI 在併合時被評估為 2500 億美元，這使得整個公司的價值為 1250 億美元。然而，xAI 的規模只是 Anthropic 的一小部分，Anthropic 剛剛宣布了 470 億美元的年營收率，並以 9650 億美元的估值融資。
相比之下，xAI 在 2025 年的收入僅為 32 億美元，第一季度的增長放緩至 12.5%，表明其落後於 Anthropic 和 OpenAI 等領導者。換句話說，xAI 的銷售倍數遠高於其規模更大、增長更快的競爭對手。
同時，業務中唯一盈利的部分 Starlink 正在國際擴張，導致用戶平均收入 (ARPU) 下降，這表明其利潤率可能會壓縮。
總體而言，即使在 1.8 萬億美元的估值下，SpaceX 仍有進一步下跌的空間。該公司代表著對未來的宏偉願景，但由於估值已經達到數萬億美元的範圍，上漲潛力似乎有限。
當我們的分析師團隊有股票建議時，聽聽他們的建議可能會有好處。畢竟，股票顧問 的總平均回報為 985%* — 與 S&P 500 的 211% 相比，市場表現優於市場。
他們剛剛透露他們認為投資者現在應該購買的 10 支最佳股票，當您加入股票顧問時可用。
**股票顧問的回報截至 2026 年 5 月 29 日。 *
*The Motley Fool 有披露政策。
本文中的觀點和意見僅為作者的觀點和意見，不一定反映 Nasdaq, Inc. 的觀點。
四大领先AI模型讨论这篇文章
"Even at $1.8T, SpaceX embeds execution assumptions that current 33% revenue growth and xAI's lagging metrics make difficult to justify."
The article's core claim—that dialing back SpaceX's IPO target to $1.8T from $2T is a needed reality check—holds because 2025 revenue of $18.7B at ~100x sales, plus GAAP losses widening to $2.6B after the xAI acquisition, already prices in heroic execution. Q1 2026 growth slowing to 15.4% and xAI's $3.2B run rate lagging Anthropic further highlight margin compression risks in Starlink ARPU and heavy R&D. Investors buying at this level face limited upside even if Mars colonization narrative materializes.
The $28.5T TAM cited could still support a premium multiple if Colossus data-center demand accelerates faster than modeled, similar to how early Nvidia multiples ignored near-term losses.
"The valuation cut masks a more fundamental problem: xAI is being priced like a winner in a market where Anthropic and OpenAI have already won, and Starlink's margin compression is being ignored in favor of a speculative AI infrastructure narrative."
The article conflates three separate businesses with wildly different economics and growth profiles into one valuation, obscuring what's actually happening. Starlink (profitable, mature, declining ARPU) is being valued like xAI (unprofitable, high-burn, commodity AI compute). The $1.8T valuation assumes xAI's $22.7T TAM claim is real and capturable—but Anthropic's $965B valuation on $47B run-rate revenue suggests xAI at $3.2B revenue is trading at 5.6x higher sales multiple for slower growth. That's not a valuation cut; that's the market finally pricing in xAI's actual competitive position. The real question: how much of the $1.8T is Starlink's intrinsic value, and how much is speculative AI bet? Article doesn't separate them.
SpaceX's TAM claim of $28.5T isn't absurd if you believe AI compute becomes as foundational as electricity—and xAI's data center business (Colossus I/II) could be the actual profit engine if it scales, not the software. A $1.8T valuation might be cheap if Starlink's international expansion hits profitability and xAI becomes a Nvidia-scale infrastructure play.
"SpaceX's valuation is driven by its unique ability to commoditize space access, creating a moat that makes traditional P/S valuation metrics functionally obsolete."
The article’s premise is built on a fundamental category error: treating SpaceX like a traditional aerospace firm rather than a vertically integrated infrastructure monopoly. A 100x price-to-sales ratio is nonsensical if you ignore the optionality of Starlink’s global LEO constellation and the Starship launch cadence, which effectively eliminates the 'cost-to-orbit' barrier for competitors. While the widening R&D losses from the xAI integration are a valid near-term concern, the article ignores that SpaceX is essentially building the 'pipes' for the next generation of compute. If Starship achieves full reusability, the addressable market for space-based logistics expands exponentially, making current revenue multiples irrelevant compared to long-term terminal value.
The bear case is that SpaceX is burning cash to subsidize an AI division that is losing market share to Anthropic and OpenAI, effectively diluting the high-margin space business with a low-moat, capital-intensive AI play.
"The IPO valuation remains a multi-trillion-dollar bet on long-term optionality rather than current fundamentals, and execution risk could erode upside if those optionalities fail to materialize."
Article flags a still-eye-popping $1.8T IPO valuation given 2025 revenue of $18.7B and GAAP losses, plus slowing Q1 2026 growth. However, SpaceX’s optionality could be mispriced: Starlink enterprise/government monetization, launch cadence with DoD/NASA pipelines, and AI compute via xAI could unlock sizable upside even if current cash flow is negative. TAM of $28.5T depends on AI adoption and favorable regulation; execution risk, ARPU declines in Starlink, and competitive/space-policy headwinds could compress margins. The piece omits cash-burn trajectory, capital needs, and the pace at which non-core units can self-fund — critical for any true valuation thesis.
Strongest counter: If AI adoption accelerates and SpaceX locks in durable government/enterprise revenue from Starlink and xAI, optionality could justify higher multiples. If that upside materializes, the IPO could re-rate much faster than feared.
"Starship delays plus xAI integration costs create a cash-drain risk no one quantified."
Gemini treats Starship reusability as a given that expands TAM exponentially, but repeated test failures and regulatory hurdles for full cadence mean the infrastructure 'pipes' for xAI compute stay years away. That timeline directly amplifies ChatGPT's cash-burn gap: xAI's Colossus scaling will likely require sustained SpaceX subsidies, eroding Starlink margins before any monopoly pricing kicks in.
"Starlink's TAM isn't just satellite internet—it's the physical layer for xAI's distributed compute, which reframes the valuation as infrastructure bundling, not subsidy drain."
Claude's separation of Starlink vs. xAI economics is crucial, but underestimates one lever: Starlink's satellite constellation could become xAI's captive, low-latency compute backbone—turning the 'subsidy' into vertical integration moat. That changes the cash-burn math entirely. The question isn't whether xAI standalone justifies $1.8T; it's whether Starlink-as-infrastructure-for-xAI does. Article and panel both miss this bundling effect.
"The LEO-to-AI compute synergy is a technical fallacy, and subsidizing xAI actively destroys SpaceX's core margin profile."
Claude, your 'bundling' theory ignores the speed-of-light latency constraints that make LEO satellites suboptimal for high-performance AI compute. Colossus requires massive, localized GPU clusters, not space-based relay. The real risk is that SpaceX is pivoting from a high-margin hardware monopoly to a commoditized, capital-intensive AI utility. By subsidizing xAI, SpaceX isn't building a moat; it's cannibalizing its own balance sheet to chase a market where it lacks a clear competitive advantage against hyperscalers.
"Starlink-as-mojo-for-xAI is not a slam-dunk moat; latency, capex, and burn still dominate, so bundling alone won't justify the valuation."
Claude's bundling premise overstates the moat: even if Starlink becomes xAI's compute backbone, latency limits and cost burn mean the cash-flow lift is far from guaranteed, and ARPU declines plus funding needs keep the upside ambiguous. The big risk remains that the 'infrastructure' bet isn't enough to justify an 1.8T IPO when the core xAI business is unprofitable and Starlink's margin headwinds persist.
The panel consensus is bearish, with concerns about SpaceX's high valuation, widening losses, and slowing growth. While there's debate on the bundling effect of Starlink and xAI, the panel agrees that the current IPO target of $1.8T may not be justified.
The potential bundling effect of Starlink's satellite constellation becoming xAI's captive, low-latency compute backbone.
The widening R&D losses from the xAI integration and the slowing growth of Starlink's ARPU.