General·Markets·AI·Glance
Sat · 1 Aug 2026

Intelligence Report

SpaceX Targets $1.75 Trillion Valuation in IPO, Nvidia Predicts $4 Trillion AI Boom

·9 min read

Executive Summary

SpaceX filed for an initial public offering on Wednesday, revealing that its Starlink satellite internet business generated over $11 billion in revenue last year and is now subsidizing deep losses in artificial intelligence, as the company projects a total addressable market of $28.5 trillion. The filing coincided with Nvidia reporting record quarterly revenue of $81.6 billion, with CEO Jensen Huang forecasting that annual spending on AI infrastructure could reach $3 to $4 trillion by 2030, a figure that sent Asian tech stocks soaring. In geopolitics, President Donald Trump has signaled a willingness to hold direct talks with Taiwan’s president, breaking a 44-year diplomatic precedent and drawing an immediate warning from Beijing, while a fragile ceasefire with Iran faces pressure from within his own political circle to be abandoned.

AI & Technology

SpaceX IPO Reveals Starlink’s Dominance and a Massive AI Bet

SpaceX filed for an initial public offering on Wednesday, revealing for the first time that its Starlink satellite internet business generated $11.39 billion in revenue last year, accounting for 61% of the company’s total sales and serving as its only profitable division. The filing, which targets a valuation of up to $1.75 trillion, showed the rocket launch unit lost $657 million in 2025 while the newly integrated AI division, which includes the social media platform X, lost $6.35 billion. Elon Musk will maintain 85% of the voting power post-IPO, and the documents include a provision granting him an additional billion shares if SpaceX establishes a colony of one million people on Mars.

The company is projecting a staggering total addressable market of $28.5 trillion, with $26.5 trillion of that tied to future AI enterprise applications—a figure critics argue is speculative. The filing also disclosed a complex $1.25 billion per month deal for compute capacity with Anthropic, a direct competitor to Musk’s own AI efforts, highlighting the desperate competition for computing power. Hours after the filing, Musk issued a public call on X for “world-class engineers and physicists” to join the SpaceXAI division, suggesting a strategic pivot to position the company as a foundational infrastructure player in AI. The immediate path forward hinges on investor appetite for a company whose profitable core business is subsidizing deep losses elsewhere and whose visionary founder retains near-total control.

Nvidia Projects Trillion-Dollar AI Spending Amid Record Earnings

Nvidia reported record quarterly revenue of $81.6 billion and net income of $58.3 billion on Wednesday, yet its shares fell 1.6% in after-hours trading as investors have grown accustomed to stellar results. CEO Jensen Huang outlined a new forecast, telling analysts that annual capital expenditures for AI infrastructure from hyperscalers like Amazon and Alphabet could reach $3 to $4 trillion by the end of the decade, a figure that dwarfs current Wall Street consensus estimates. He also identified a “brand new” $200 billion market opportunity for the company in AI-specific central processing units through its new Vera product, directly challenging incumbents like Intel and AMD.

The company’s bullish outlook provided a global boost to tech sentiment, most notably in South Korea where the benchmark KOSPI index soared more than 8%. The rally followed a last-minute deal to avert a crippling strike at Samsung Electronics, a critical memory chip supplier for the AI industry. However, the market’s muted reaction to Nvidia’s own financials underscores the immense pressure on the $5.3 trillion company to continually surpass sky-high expectations. The company disputed any notion of a slowdown, with Huang stating “demand has gone parabolic.” The immediate test is whether Nvidia’s forecast for next-quarter revenue of approximately $91 billion can reignite investor enthusiasm.

Anthropic Nears First Profit While Negotiating for Microsoft AI Chips

Anthropic is on track to post $10.9 billion in revenue and an operating profit of $559 million for the quarter ending in June, according to internal figures confirmed by CNBC, which would mark the AI startup’s first profitable quarter since its founding. Concurrently, the company is in discussions with Microsoft to use its custom Maia 200 AI chips, a potential deal that follows Microsoft’s $5 billion investment in Anthropic last November. The reported financial surge, driven by the rising popularity of its Claude chatbot, positions Anthropic as a formidable competitor to OpenAI, with its valuation reportedly targeted at $900 billion in a current funding round.

However, Anthropic has told investors it does not expect to remain profitable in subsequent quarters, citing massive planned expenditures on computing power—a pressure underscored by its separate $1.25 billion monthly compute deal with SpaceX. The potential chip deal with Microsoft highlights the intensifying infrastructure race in AI, as Microsoft lags behind Amazon and Google in supplying custom AI silicon to cloud clients. For Anthropic, which has historically relied on Nvidia GPUs and cloud services from its other investors, the arrangement would diversify its supply chain. The company’s financial momentum comes despite a high-profile clash with the U.S. government, which labeled it a “supply chain risk” after it refused to remove AI safeguards for military applications.

Tesla Launches Full Self-Driving in China After Years of Delay

Tesla announced Thursday that its Full Self-Driving supervised driver-assistance system is now available for purchase in China, ending a multi-year regulatory standoff. The technology is listed as an add-on for the Model 3 on Tesla’s Chinese website for a one-time fee of 64,000 yuan ($9,410). The launch comes one week after Tesla CEO Elon Musk visited Beijing as part of a U.S. business delegation accompanying President Donald Trump.

The arrival of Tesla’s FSD intensifies competition in a market where domestic rivals like BYD, Xpeng, and Huawei have already rolled out advanced systems and have a head start in navigating China’s complex urban environments. Tesla’s system, which requires active driver supervision, had faced significant regulatory delays, with Musk repeatedly predicting its imminent approval since 2024. It is unclear how quickly Chinese consumers will adopt the costly system given established local alternatives. The announcement also coincides with a broader surge in Chinese EV exports, as domestic automaker Chery Automobile projected its overseas EV sales could exceed 1 million units this year.

Samsung Averts Strike With Tentative Deal on AI Chip Profits

Samsung Electronics reached a last-minute tentative agreement with its largest union late Wednesday, suspending a planned 18-day strike by nearly 48,000 workers that was set to begin Thursday. The deal, brokered by South Korea’s labor minister, includes a revised bonus system where Samsung will allocate 10.5% of its annual operating profit to employee bonuses and abolish a previous cap, though the union had demanded 15%. Union members will vote on the agreement from May 22 to 27, meaning the immediate threat to production is delayed but not fully resolved.

The stakes were global, as Samsung controls roughly 36% of the DRAM market. Analysts had warned an 18-day walkout could have disrupted 4% of global DRAM supply, exacerbating shortages amid already strained AI-driven demand. For South Korea, where Samsung accounts for about a fifth of economic output, a prolonged strike was estimated to cost the company up to $660 million per day. The agreement represents a significant concession from the union, which had initially demanded a larger share of profits. The outcome will be closely watched by rivals SK Hynix and Micron as it sets a precedent for labor relations in the semiconductor industry during an AI boom.

Geopolitics & Security

Trump Offers Direct Talks With Taiwan, Testing China After Summit

President Donald Trump has signaled a willingness to hold direct talks with Taiwan’s President William Lai Ching-te, a move that would break more than four decades of diplomatic protocol and has drawn an immediate warning from Beijing. The offer, made twice since Trump’s state visit to China last week, comes as the White House weighs a $14 billion arms package for Taiwan, which Trump has called a “very good negotiating chip.” China’s foreign ministry reiterated its “consistent, clear and firm” opposition to any official U.S.-Taiwan exchanges or arms sales, urging Washington to handle the Taiwan question with “actual prudence.”

Taiwan’s Foreign Ministry said Lai would be “happy to discuss these matters with President Trump.” The U.S. and Taiwanese presidents have not spoken directly since Washington shifted diplomatic recognition from Taipei to Beijing in 1979. The potential arms sale and talk of direct dialogue represent a significant test of the understandings reached during Trump’s summit with Chinese President Xi Jinping last week. The tension is already affecting military diplomacy, with Beijing holding up a proposed summer visit by the Pentagon’s top policy official until Trump decides on the arms package. Experts suspect China will use future defense official visits as leverage to pressure the administration.

Russia and NATO Conduct Drills Amid Rising Airspace Tensions

Russia and Belarus conducted a joint strategic nuclear exercise on Thursday, with Presidents Vladimir Putin and Alexander Lukashenko participating via video link for the first time. The Russian Defense Ministry stated that nuclear munitions were delivered to field storage facilities in Belarus for the drills, prompting Ukraine to announce enhanced security measures along its northern border. Concurrently, NATO air forces were scrambled over Latvia after at least one drone was detected in its airspace, the latest in a series of Baltic incidents.

Separately, the United Kingdom released footage on Wednesday that it said showed two Russian fighter jets intercepting a British RAF Rivet Joint spy plane over the Black Sea, with one jet flying as close as six meters. In the Pacific, the U.S. military will deploy its Typhon midrange missile launcher and HIMARS rocket system to Japan’s Kanoya Airbase next month for joint exercises with Japanese forces, a move sources said is aimed at deterring China. These simultaneous military activities reflect a global landscape of heightened strategic posturing, with daily friction points risking miscalculation.

Trump and Netanyahu Clash Over Iran as Ceasefire Talks Continue

President Donald Trump and Israeli Prime Minister Benjamin Netanyahu are at odds over whether to pursue a ceasefire with Iran or resume military strikes, according to multiple reports. A tense phone call on Tuesday left Netanyahu furious, with one U.S. source telling Axios the Israeli leader’s “hair was on fire” after Trump delayed a planned “very major attack” to give diplomacy more time. Trump has since publicly stated that Netanyahu “will do whatever I want him to do,” a claim that underscores the significant pressure the U.S. is applying on its closest Middle Eastern ally to stand down.

The friction exposes a strategic divergence. Netanyahu’s government is pushing to restart hostilities to further degrade Iran’s military, while Trump has oscillated between threats and diplomacy, telling reporters this week the choice was between “a deal” or doing “some things that are a little bit nasty.” This mixed messaging coincides with Iran confirming it is reviewing an updated U.S. proposal. Analysts note that Israel’s ability to act alone is severely limited, and the political cost for Netanyahu is high. The immediate future hinges on Tehran’s response and whether Trump’s assertion of control over Netanyahu holds.

Ebola Outbreak in Congo Escalates as Mob Torches Treatment Center

A mob in eastern Democratic Republic of Congo set fire to an Ebola treatment center in Rwampara on Thursday, destroying isolation tents and forcing health workers to flee, after authorities prevented them from retrieving the body of a young man suspected of dying from the virus. The attack, which injured a healthcare worker, underscores the severe public mistrust hampering the response to a rapidly spreading outbreak of the rare Bundibugyo strain, which has killed more than 130 people in Ituri province. The World Health Organization has declared the outbreak a “public health emergency of international concern.”

Experts warn the 600 confirmed cases represent only “the top of the iceberg” and that the true number is likely far higher, exacerbated by the lack of a specific vaccine for this strain. The violence reflects a dangerous collision between necessary public health protocols and deep-seated local customs and beliefs. The combination of armed conflict, community resistance, and a vaccine gap threatens to let the outbreak spiral further out of control in one of the world’s most vulnerable regions.

From the Timeline

Media Bias and Political Commentary Spark Debate

A significant thread focused on perceived media bias and political hypocrisy. @elonmusk amplified criticism of British politicians and Reuters for their coverage of a specific crime versus the George Floyd case, framing it as a deliberate narrative. This sentiment of institutional failure extended to domestic politics, where @paulg speculated that political silencing tactics might be linked to hidden information in the Epstein files. Meanwhile, @Noahpinion shared polling data suggesting deep unpopularity for progressive ideology within the Democratic Party.

The Transformative and Disruptive Power of AI

The potential and immediate impact of AI was a dominant theme, discussed from multiple angles. @pmarca enthusiastically shared a prediction that AI will achieve superhuman performance in mathematics within a few years, leading to “stratospheric” civilizational payoffs. @garrytan framed the current moment as the “Apple II moment for AI,” urging builders to create the personal AI tools of the future. In a more immediate business context, @garrytan also highlighted a Cloudflare CEO’s view that AI is poised to displace “measurers” (analysts, middle managers) rather than “builders or sellers,” forecasting widespread workforce restructuring even at high-growth companies.

The Evolving AI Product and Developer Landscape

Beyond speculation, thought leaders shared concrete updates and debated AI’s commercial trajectory. @sama announced a new update to OpenAI’s Codex, featuring “Appshots” for better context integration. @fchollet predicted a fundamental shift in computing, where “Apps become services and UIs become text boxes.” A debate emerged around where value accrues, with @ClementDelangue noting that several AI investors believe “frontier tokens are capturing the majority of the economic value,” overshadowing older models. @patrickc pointed to a new platform, Index, as a solution for content owners to monetize AI agent usage, arguing that agents can solve the micropayment problem that stumped humans due to decision fatigue.

Technology, Infrastructure, and Business Momentum

Several posts highlighted bullish signals and strategic moves across tech sectors. @zerohedge noted SpaceX’s pledge in its S-1 to cover power grid upgrade costs for its data centers, suggesting a new standard for hyperscalers. @tobi celebrated Shopify’s resilience against short sellers, sharing an analysis that called a past short report one of the “worst calls of all time.” In enterprise software, @chamath detailed his firm’s work building an AI-powered “operating system” for regulated industries like pharma, aiming to streamline documentation and boost R&D efficiency.

Societal Observations: From Status Symbols to Urban Policy

Commentary extended to broader cultural and societal trends. @levelsio sparked a thread on the decline of traditional luxury brands, arguing that their exclusivity models are meant for “poor people to fake they’re rich” and that billionaires now find them inconvenient. On urban policy, @pmarca shared a call to reframe the homelessness crisis around personal responsibility and compassion, criticizing existing political and NGO approaches. In a different civic context, @garrytan endorsed Flock Safety’s license plate reader technology as a tool to make cities like San Francisco safer.

Methodology

Total Articles1720
Used Articles1436
Total Sources112
Used Sources56

Newsletter

In your inbox every morning, 5 AM ET.