首页  >>  来自播客: Solving The Money Problem 更新   反馈  

Solving The Money Problem - WILD: SpaceX $3 Trillion + Tesla Merger Theory

发布时间:   原节目
以下是该内容的中文翻译: 这份详细总结汇集了转录内容中所有关于SpaceX、特斯拉及相关业务的重要新闻、预测和观点: **SpaceX整体财务预测与估值:** * **当季营收:** SpaceX本季度营收达到78亿美元,同比增长92%。 * **近期年化经常性收入(ARR):** 预计到今年12月底,年化经常性收入(ARR)将达到1000亿美元。 * **长期营收目标:** 埃隆·马斯克已将万亿美元*年营收*(非年化经常性收入)目标从2031年提前到2030年,并表示在2029年实现该目标存在“非零概率”。 * **万亿美元目标的背景:** 世界历史上没有一家公司曾实现万亿美元年营收。尽管如此,SpaceX在2030年实现这一目标的路径被描述为“合理”、“可行”,且“不需要奇迹”。 * **分析师预估:** 摩根士丹利对SpaceX 2030年的营收预估为3250亿美元(相比之下,马斯克的目标是1万亿美元)。 * **当前估值:** SpaceX当前估值约为1.4万亿美元。相较于其IPO后峰值,估值下跌了40-50%,这在科技股中很常见。 * **基于模型的估值(Grok):** 一个内部(现已对外公开)估值模型表明,在2030年代初中期(基本情况),或在2030年正负两年内(牛市情况),SpaceX有一条可信的路径能够实现万亿美元年营收。 * **相较于特斯拉的价值:** 该模型表明SpaceX目前的价值约为特斯拉的两倍(2026年两者合并公允价值中,SpaceX约占66%,特斯拉约占34%)。如果SpaceX市值达到3万亿美元,而特斯拉市值保持不变,则SpaceX将占总值的约70%,特斯拉约占30%。 **实现万亿美元营收目标的关键驱动因素:** 1. **人工智能算力:** * **算力容量:** SpaceX计划在2026年底前达到2吉瓦(GW)的算力,并到2027年底前达到10吉瓦。 * **营收潜力:** 埃隆表示,算力的现货价格为每瓦30-50美元。若以2吉瓦、每瓦50美元的算力运行,仅此一项就能达到1000亿美元的年化经常性收入(ARR)。 * **云服务协议:** 今年下半年已签署了67亿美元的云服务协议。 * **租赁业务:** SpaceX已将大量算力租赁给Anthropic。问题在于他们是会建立自己的前沿模型还是继续租赁,值得注意的是,GPU租赁业务通常估值倍数较低(例如CoreWeave)。 * **地面与轨道:** 预计AI算力对万亿美元目标的主要贡献将来自地面而非轨道。 2. **星链(Starlink):** * **用户增长:** 目前拥有1200万用户,同比增长一倍(原约为600万)。按当前增长速度推算,用户运行速率可达2400万。 * **当前营收:** 星链营收增长66%,达到43亿美元(季度营收,或为本季度78亿美元总营收的一部分)。 * **EBITDA:** 本季度连接业务产生了26亿美元的调整后EBITDA。 * **ARPU:** 平均每用户营收(ARPU)为每月66美元,环比增长20%。 * **企业星链:** 正迅速接近星链总营收的一半,预计未来将成为主要部分。 * **直连手机(DTC):** * 格温·肖特韦尔(Gwen Shotwell)暗示DTC将占据6000亿美元移动通信总潜在市场(TAM)的“很大一部分”。 * 一颗星链卫星可以服务比家庭宽带用户(4K视频、下载)多许多倍的手机用户(语音、轻度互联网)。 * 用户增长预计将“迅速逼近并超过1亿用户,比任何人预期的都要快得多”,尽管平均每用户营收(ARPU)较低,但大部分用户将倾向于手机服务。 * 目前每季度新增200万消费者用户,与手机集成后,每季度可能加速增至400-500万。仅美国就有4亿移动用户。 * **市值潜力(仅星链):** 可能在18-24个月内达到1万亿美元市值,每年产生400亿美元营收和300亿美元自由现金流(基于30倍市盈率)。 * **马斯克对星链估值的看法:** 针对星链万亿美元估值的说法,马斯克表示“远不止于此”。他认为,由于人工智能和机器人技术的发展,带宽需求将“数量级”增长(100倍或更多)。他预测,星链在中国以外的市场份额至少可达25%(每年营收超过5000亿美元),长期来看,甚至可能承载超过50%的互联网流量(每年超过1万亿美元)。 * **竞争(或缺乏竞争):** 有人认为,由于SpaceX在发射技术(星舰)方面拥有10年以上的领先优势,以及其大规模的卫星部署能力,其他公司无法有效与星链竞争。 **星舰的关键作用:** * **V3卫星:** 星舰对于部署新的V3星链卫星至关重要,V3卫星提供V2卫星10倍的带宽。 * **部署能力:** 猎鹰9号每次可发射27颗V2卫星(网络容量约2.6太比特/秒)。星舰每次可部署60颗V3卫星(约60太比特/秒),每次发射提供的容量是猎鹰9号的20多倍。 * **近期测试:** 最近一次星舰测试证实隔热罩正常工作,并成功部署了20颗V3卫星进行测试,并建立了连接。 * **下一次测试:** 下一次星舰飞行将首次将活跃的V3卫星部署到星座中以提供服务。 **人工智能前沿模型(Grok与Cursor):** * **Grok Build 4.5:** 导致7月份令牌使用量立即增加了两倍,这表明其在“单位智能成本方面具有最佳性价比”。 * **Cursor:** 原本预计年底前将从30亿美元增长到100亿美元,在Grok的加持下,可能达到100-200亿美元。 * 这些人工智能模型业务的估值倍数预计将远高于数据中心租赁业务。 **实现万亿美元营收的“大胆预测”:** * **路径1:通过擎天柱(Optimus)实现SpaceX-特斯拉合并:** 埃隆的动机是推动SpaceX估值足够高(例如3万亿美元)以收购特斯拉,从而获得“无可争议的控制权”。这将伴随着擎天柱机器人在地球和太阳系中接管服务/体力劳动。 * **路径2:TerraFab取代台积电:** SpaceX通过其“TerraFab”项目取代台积电成为主要的芯片制造商。台积电的年营收接近1000亿美元,而TerraFab可能达到其10-20倍,与英伟达合作成为“美国台积电”。这将解决英伟达对台湾芯片生产的依赖。埃隆曾表示,TerraFab的目标是生产比世界其他地区加起来还要多的芯片。 **埃隆·马斯克的领导力与哲学:** * **冒险精神:** 他因其“对创新的不羁热情”和愿意进行“高风险但极其重要的投资”而受到赞扬,他将星链“令人难以置信的巨额现金机器”投入到TerraFab、AI数据中心和AI模型等项目中,而不是用于股票回购或分红。 * **硬件专长:** 他认为建设数据中心“不是什么高深学问”(相较于建造火箭),这突显了SpaceX在快速扩展硬件密集型业务方面的优势。 **潜在的协同效应与未来发展:** * **特斯拉整合:** 未来销售的每一辆特斯拉都将内置星链,创建Wi-Fi网络,将手机连接到特斯拉(例如自动驾驶出租车),并直接连接到下一代星链网络。 **阻力/质疑(可能被证明是错误的论点):** * **AI算力需求:** 担忧令牌和算力需求可能无法持续,或者本地部署解决方案、桌面应用和开源模型可能会抑制需求。 * **星链市场定位:** 认为星链仅适用于农村地区,无法在城市地区与光纤/有线宽带竞争的论点(尽管提出了特斯拉集成作为反驳)。 **个人投资立场:** * 演讲者正在“积极”地在1.5万亿美元左右的估值水平上建立SpaceX头寸,并认为其长期潜力远高于特斯拉,尽管对特斯拉仍持看涨态度。

This detailed summary compiles every significant news item, projection, and opinion discussed in the provided transcription regarding SpaceX, Tesla, and related ventures: **Overall SpaceX Financial Projections & Valuation:** * **Current Revenue:** SpaceX posted $7.8 billion in revenue this quarter, a 92% year-over-year increase. * **Near-Term ARR:** Expects to hit $100 billion in Annual Recurring Revenue (ARR) by the end of December this year. * **Long-Term Revenue Target:** Elon Musk has moved the target for $1 trillion in *annual revenue* (not ARR) from 2031 to 2030, with a "non-zero chance" of hitting it in 2029. * **Context for $1T Target:** No company in world history has ever reached $1 trillion in annual revenue. Despite this, the path to SpaceX achieving this by 2030 is described as "reasonable," "viable," and requiring "no miracles." * **Analyst Estimates:** Morgan Stanley's 2030 revenue estimate for SpaceX is $325 billion (compared to Musk's $1T). * **Current Valuation:** SpaceX is currently valued around $1.4 trillion. It's seen a 40-50% drop from its peak post-IPO, which is typical for tech stocks. * **Model-Based Valuation (Grok):** An internal (now external) valuation model suggests a credible path to $1 trillion annual revenue by early-mid 2030s (base case) or comfortably within 2030 +/- 2 years (bull case). * **Value Relative to Tesla:** The model suggests SpaceX is currently worth roughly double Tesla (SpaceX ~66%, Tesla ~34% of combined fair value in 2026). If SpaceX reached $3 trillion market cap while Tesla stayed at its current market cap, SpaceX would represent ~70% and Tesla ~30%. **Key Drivers for $1 Trillion Revenue Target:** 1. **AI Compute:** * **Compute Capacity:** SpaceX plans to hit 2 gigawatts (GW) of compute by the end of 2026 and up to 10 GW by the end of 2027. * **Revenue Generation:** Elon stated the spot price for compute is $30-$50 per watt. At 2 GW running at $50/watt, this alone could hit $100 billion ARR. * **Cloud Service Deals:** Inked $6.7 billion in cloud service deals in the second half of this year. * **Rental Business:** SpaceX has rented a huge block of compute to Anthropic. There's a question whether they will build their own frontier model or continue renting, noting that GPU rental businesses tend to trade at low multiples (e.g., CoreWeave). * **Terrestrial vs. Orbital:** The primary AI compute contribution to the $1T target is expected to be terrestrial, not orbital. 2. **Starlink:** * **Subscriber Growth:** Currently has 12 million subscribers, which doubled year over year (from ~6M). Extrapolating current growth, it could reach a 24 million subscriber run rate. * **Current Revenue:** Starlink revenues grew 66% to $4.3 billion (quarterly, or as a segment of the $7.8B total for the quarter). * **EBITDA:** Generated $2.6 billion in adjusted EBITDA in the quarter from connectivity. * **ARPU:** Average Revenue Per User (ARPU) is $66 per month, growing 20% quarter-over-quarter. * **Enterprise Starlink:** Fast approaching half of all Starlink revenues and is expected to become the majority in the future. * **Direct-to-Cell (DTC):** * Gwen Shotwell implied DTC will take a "big chunk" of the $600 billion mobile telephony Total Addressable Market (TAM). * One Starlink satellite can serve many times more cell customers (voice, light internet) than home broadband customers (4K video, downloads). * Subscriber growth is expected to "rocket toward and then exceed 100 million customers much sooner than anyone expects," with the majority skewing towards cell service, despite lower ARPU. * Currently adding 2 million consumer subscribers per quarter, which could accelerate to 4-5 million per quarter with handset integration. There are 400 million mobile subs in the US alone. * **Market Cap Potential (Starlink Alone):** Could be a $1 trillion market cap within 18-24 months, generating $40 billion revenue and $30 billion free cash flow annually (based on a 30x multiple). * **Musk's View on Starlink Valuation:** Responding to the $1T Starlink valuation idea, Musk stated it would be "far more than that." He believes bandwidth demand will increase "orders of magnitude" (100x or more) due to AI and robotics. He predicts Starlink could reach at least 25% market share outside of China (over $0.5 trillion revenue/year) and potentially carry over 50% of internet traffic long-term (over $1 trillion/year). * **Competition (or lack thereof):** It's argued that no other company can effectively compete with Starlink due to SpaceX's 10+ year lead in launch technology (Starship) and the massive scale of their own satellite deployment. **Starship's Critical Role:** * **V3 Satellites:** Starship is essential for deploying the new V3 Starlink satellites, which offer 10x the bandwidth of V2 satellites. * **Deployment Capacity:** Falcon 9 launches 27 V2 satellites (~2.6 terabits/second network capacity). Starship can deploy 60 V3 satellites (~60 terabits/second), providing over 20 times more capacity per launch. * **Recent Test:** The last Starship test confirmed the heat shield worked and successfully deployed 20 V3 satellites as a test, making connections. * **Next Test:** The next Starship flight will be the first to deploy active V3 satellites into the constellation for service. **AI Frontier Models (Grok & Cursor):** * **Grok Build 4.5:** Resulted in an immediate tripling in token usage in July, demonstrating "best value for money in terms of cost per intelligence." * **Cursor:** Was already on track to grow from $3 billion to $10 billion by year-end, potentially reaching $10-20 billion with Grok. * These AI model businesses are expected to trade at much higher multiples than the data center rental business. **"Hot Takes" on Achieving $1 Trillion Revenue:** * **Path 1: SpaceX-Tesla Merger via Optimus:** Elon's motivation to drive SpaceX valuation high enough (e.g., $3 trillion) to buy Tesla, thereby gaining "unquestionable control." This would happen with Optimus robots taking over service/physical labor on Earth and in the solar system. * **Path 2: TerraFab Replaces TSMC:** SpaceX, through its "TerraFab" project, replaces TSMC as the dominant chip manufacturer. TSMC's revenue is nearly $100 billion/year, and TerraFab could be 10x-20x that, becoming an "American TSMC" in partnership with Nvidia. This would address Nvidia's dependency on Taiwan for chip production. Elon's stated goal for TerraFab is to produce more chips than the rest of the world combined. **Elon Musk's Leadership & Philosophy:** * **Risk-Taking:** Praised for his "unbridled enthusiasm for innovation" and willingness to take "highly risky, but highly important investments" by plowing Starlink's "unbelievable juggernaut cash machine" into ventures like TerraFab, AI data centers, and AI models, rather than share buybacks or dividends. * **Hardware Expertise:** His belief that building data centers is "not rocket science" (compared to building rockets) highlights SpaceX's advantage in quickly scaling hardware-intensive operations. **Potential Synergies & Future Developments:** * **Tesla Integration:** Every future Tesla sold will have Starlink built in, creating Wi-Fi networks that connect phones to Teslas (e.g., robo-taxis) and directly to the next-generation Starlink network. **Headwinds/Skepticism (Arguments that may prove wrong):** * **AI Compute Demand:** Concerns that demand for tokens and compute might not keep up, or that on-prem solutions, desktops, and open-source models could mute demand. * **Starlink Market Niche:** The argument that Starlink is only for rural areas and won't compete with fiber/cable in urban areas (though Tesla integration is proposed as a counter). **Personal Investment Stance:** * The speaker is "aggressively" building their SpaceX position around the $1.5 trillion mark and views its long-term potential as significantly higher than Tesla's, though still bullish on Tesla.