On August 5, 2026, this originally ordinary summer trading day was marked by the market repeatedly afterward: the Nikkei 225 index opened above 64,000 points, rising about 1% at the start of trading, while the Korean KOSPI index opened about 3.8% higher, reaching around 6,600 points. SK Hynix rose over 7%, and Samsung Electronics increased by about 5% (according to a single source). Coupled with the positive technical and political progress in negotiations between Iran and Oman regarding the Strait of Hormuz, and the circulation of a potential "U.S.-Iran agreement" draft among various parties (according to a single source), global risk appetite seemed to jump to a higher level all at once. On the same day, Palantir announced Q2 2026 revenue of about $1.94 billion, a year-on-year increase of approximately 93% (according to a single source), with its stock price rising about 29%-30% in a single day after the earnings report, closing at around $162.66, with a market capitalization nearing $389.95 billion; on the other side, Microsoft, Meta, Oracle, Amazon, and Alphabet were revealed to collectively have committed about $1.09 trillion in future AI data center leasing payments, with lease liabilities currently at about $285 billion (according to a single source). SpaceX reported Q2 revenues of about $7.8 billion, a net loss of about $541 million, AI business revenues of about $2.56 billion, and total capital expenditures of approximately $18.369 billion (according to a single source), and in the same communication, intensified its "full-stack AI" narrative with Grok iteration and multi-GW level computing power planning. All these pieces together formed a narrative that has been repeatedly digested: AI capital expenditures are entering a trillion-dollar "super cycle," while the real suspense lies in whether this is truly the starting point of a long-term boom, or rather a prelude that will be defined years later as a new round of overcapacity and valuation bubble.
Japanese and South Korean Stock Indices Open Higher: Risk Appetite Warming Up
On the same day when the narrative of trillion-dollar AI investment was amplified, Tokyo and Seoul's markets provided immediate feedback. On August 5, 2026, the Nikkei 225 index opened above 64,000 points, rising about 1% initially, and the simultaneous high opening of the index was interpreted as a rapid follow-up to the reversal of global tech stock sentiment. On the same day, the KOSPI index opened about 3.8% higher, reaching around 6,600 points (according to a single source), appearing particularly aggressive among major developed markets, as if declaring with a "gap up" stance that the previously suppressed risk appetite was being released.
More indicative were the sector details. Among KOSPI constituent stocks, SK Hynix rose over 7%, and Samsung Electronics increased by about 5% (according to a single source), both regarded as the hardware hub for AI storage and general chips, interpreted by the market as pricing in future demand early—when upstream tech giants roll out hefty computing power and data center plans, Asian hardware manufacturers' stock prices first provided the answer to "who will supply." On the geopolitical level, the Iranian Foreign Ministry spokesperson stated that negotiations with Oman regarding the Strait of Hormuz had made "positive progress" on both technical and political levels (according to a single source), while the Qatari Foreign Ministry said that the draft concerning the potential "U.S.-Iran agreement" was circulating among various parties, but there was no set agenda for direct U.S.-Iran negotiations (according to a single source); although these signals still carry uncertainty regarding actual implementation, they were enough to mitigate some tail risk imaginations and provide a relatively safe emotional release window for funds searching for high beta targets in the AI hardware and export chain in the Japanese and South Korean stock markets.
Palantir Soars 30% Overnight
On the same day that the Japanese and South Korean indices opened strong amid the narrative of AI hardware, another key storyline in AI was ignited in the U.S. stock market. Palantir released its Q2 2026 financial report before trading: revenue of about $1.94 billion, a year-on-year surge of about 93% (according to a single source), and the market generally linked this "nearly doubling" revenue rhythm directly to its expansion of orders for government and corporate AI solutions. In the subsequent trading day, funds voted with their feet—Palantir's stock price soared about 29%-30% in a single day, closing at approximately $162.66, which lifted its market cap to about $389.95 billion, making it a company referred to by some media as the "leader in AI software," completing a new round of valuation re-pricing in just a few hours.
This re-pricing was not an isolated emotional spur, but resonated with the macro backdrop of warming global risk appetite and escalating AI narratives that day: the Japanese and South Korean markets endorsed the hardware end with the rises of SK Hynix and Samsung Electronics, while U.S. investors were betting on Palantir as an alternative path of "software consuming AI." The former corresponds to the capital expenditure cycle of foundries, storage, and computing power equipment, while the latter refers to a high-growth revenue curve centered around subscription contracts and project deliveries; while the hardware sector is discussing capacity and supply, Palantir is focused on orders and use cases. On a valuation level, these two tracks seem to diverge but are actually mutually reinforcing, jointly forming the narrative of "dual-driven AI software and hardware" capital that day.
Trillion-Dollar Leasing Commitment: Arms Race for AI Data Centers
If Palantir represents the "light asset" AI software story, then another set of numbers disclosed on the same day brought the narrative back to concrete and steel. Five giants—Microsoft, Meta, Oracle, Amazon, and Alphabet—were reported to have collectively committed about $1.09 trillion in future AI data center leasing payments, with these contracts pointing not to office buildings and shopping malls, but to facilities and computing power infrastructure specifically built for AI and cloud computing. According to AiCoin data, they currently have confirmed leasing liabilities of only about $285 billion, far below the total amount pledged, indicating that this investment has already signed a "military order", but is still just the tip of the iceberg in their financial reports.
This action of locking in the data centers, power, and racks for the next several years all at once was quickly labeled by the market as an "AI computing power arms race": while chips are still not fully in place and use cases are still being explored, infrastructure is already arranged on a trillion-dollar scale. Optimists believe that as long as AI application layer revenues can rise quickly like Palantir's in the coming years, these heavy leases and capital expenditures will ultimately be covered by a new generation of software and services; but pessimists worry that if AI demand growth falls short of expectations, the already signed contracts could solidify overcapacity on the financial reports, turning what is praised today as forward-looking "armament" into fixed costs that are difficult to retreat from tomorrow.
SpaceX Bets on AI: Increasing Computing Power Amid Losses
In the same day’s financial report, SpaceX presented a balance sheet pulled in two directions: one side showed Q2 2026 revenue of about $7.8 billion, corresponding to a net loss of about $541 million; the other side revealed capital expenditures of about $18.369 billion, invested simultaneously in rockets, satellites, and AI infrastructure. According to a single source, the AI business alone reported quarterly revenues exceeding $2.56 billion, nearing one-third of total revenues, indicating that this company, once seen as a "pure aerospace company," is shifting its revenue and asset structure toward being an "AI computing power and model supplier."
To prove this is not just a one-off story, Musk outlined an almost non-stop Grok iteration timeline during the earnings call: 4.6 is expected to launch next week, 4.7 will follow in a few weeks, and Grok 5 is planned to be launched this year, claiming it will directly ingest SpaceX's "full data" for training. He also provided a numerical target for the expansion of computing power—according to a single source, SpaceX aims to exceed 2GW by the end of 2026 and approach 10GW by the end of 2027. The recently launched Grok 4.5, which emphasizes code and financial legal task capabilities, has not yet been fully absorbed by the market, yet a new round of models and data centers is already on the way. The huge capital expenditures combined with ongoing losses make SpaceX appear to be more "all-in" than traditional cloud companies in this AI arms race, tying financial leverage and technological bets together, becoming one of the most closely watched yet risk-laden participants in the coming years.
Opportunities and Concerns in the AI Super Cycle
On the same trading day, the Nikkei 225 and KOSPI opened strongly against the backdrop of slightly easing tensions in the Middle East, Palantir magnified the imagination of "AI software premium" with nearly double year-on-year Q2 revenue and a near 30% single-day stock price increase, while the gap between Microsoft, Meta, Oracle, Amazon, Alphabet’s total commitment of about $1.09 trillion in future AI data center leasing and their approximately $285 billion in liabilities represents an invisible boulder that will weigh down financial reports in the coming years during this super cycle; at the same time, SpaceX, while still reporting a loss of about $541 million in Q2, bets nearly a third of its revenue on AI business and plans to expand its computing power from 2GW to nearly 10GW, yet lacks transparency regarding the specific structure of its AI business, Palantir's revenue structure, and the itemized leasing commitments of each tech giant, which makes the return cycle and end demand of the AI arms race resemble a black box gamble. Next, whether the situation in the Strait of Hormuz and potential U.S.-Iran agreements can continue to lower risk premiums, whether the macro interest rate environment is genuinely reaching an inflection point, and whether cash flows from the AI application layer can continuously digest this round of infrastructure investment will determine whether this super cycle transforms into a long-term technology capital support curve or leads to a collective correction of valuations for tech stocks and crypto assets when demand realization falls short of expectations. For investors, this means that at the noisiest time in the AI super narrative, what truly needs to be closely watched are whether the three curves of risk appetite, capital cost, and real demand can continue to align.
Join our community to discuss and become stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
