After the earnings reports of SpaceX and AMD were released, both saw a significant drop in after-hours trading, which may reveal the trend of the next phase of the U.S. stock market

Last night, the U.S. stock market experienced a long-awaited broad rally.

However, the market sentiment after the close quickly returned to a tense state, pulled back by two heavyweight earnings reports—SpaceX and AMD released their latest quarterly results almost simultaneously after hours. The numbers themselves had their highlights, but the market's reaction was surprisingly consistent: both stocks fell.

SpaceX dropped over 7% after hours, while AMD fell over 9%. Both companies exceeded revenue expectations, and their guidance wasn't bad, so why did investors choose to vote with their feet?

The answer lies in the same reason—the market's "passing line" for AI narratives has been raised to an almost harsh height.

1. SpaceX: Rapid Growth, But Burning Cash Even Faster

SpaceX released its first quarterly earnings report since going public.

Revenue was $7.8 billion, exceeding market expectations. As a "debut," this number is impressive. However, the problem lies in capital expenditures.

The earnings report showed that SpaceX's capital expenditures in the second quarter reached $18.369 billion, with about $16 billion invested in AI computing infrastructure. This figure far exceeded analysts' previous expectations of $13.2 billion.

$16 billion is almost equivalent to the total revenue of SpaceX for two quarters. In other words, for every dollar the company earns, it is burning more than $2 on AI infrastructure.

Market concerns about this "burn rate" were directly reflected in the stock price. Despite Musk maintaining an extremely aggressive long-term vision during the earnings call—bringing the company's annual revenue target of $1 trillion forward from 2031 to 2030—these long-term narratives clearly did not soothe investors' worries about short-term cash returns.

Musk also emphasized that the market severely underestimates the potential of Starlink. He stated that within the next decade, Starlink is expected to provide major global internet services in most countries where operations are permitted. SpaceX's long-term growth path is anchored on three pillars: the expansion of Starlink, the commercialization of Starship, and the return on investment in AI infrastructure.

But a harsh reality is: if capital investment cannot quickly translate into quantifiable revenue growth, the trillion-dollar revenue target is just a number on paper.

What further unsettles the market is that after the earnings report, the company will face the unlocking of 910 million shares—this number even exceeds the current free float. Such a large-scale release of shares will almost inevitably create heavy supply pressure on the stock price.

In short, SpaceX's situation is awkward: long-term narratives are fully loaded, short-term cash burn is excessive, and the share structure is poor. It possesses almost all the elements of a "great company," but at this current moment, it is still some distance from being an "ideal investment target."

2. AMD: Numbers Exceed Expectations Across the Board, But the Market Wants "Significantly Exceeding Expectations"

AMD's earnings report, from a numerical standpoint, is nearly perfect.

Data center revenue was $6.72 billion, a year-over-year increase of 107% and a quarter-over-quarter increase of about 15.6%, serving as the core engine driving overall performance. Total revenue was $11.54 billion, a year-over-year increase of 50%, setting a historical high and exceeding market expectations of $11.28 billion. Adjusted earnings per share were $1.66, surpassing analysts' expectations of $1.62.

The guidance is also positive: third-quarter revenue is expected to be around $13 billion, higher than the market expectation of $12.52 billion; adjusted gross margin is about 56%.

In terms of AI business, AMD is accelerating its challenge to NVIDIA's dominance. CEO Lisa Su announced that the second-generation Helios AI servers, equipped with MI455X accelerators and TSMC's "Venice" processors, have been fully produced and are expected to start shipping in the coming months. Reports indicate that AMD will begin delivering related AI servers to Meta and OpenAI by the end of this quarter.

Almost all numbers indicate a very favorable situation, yet the stock price plummeted over 9% after hours.

Why? Because AMD's stock price has already doubled this year, and the valuation premium given by the market has reached an extreme level—DCF fair value is only about $130, while the actual stock price is as high as four to five hundred dollars. In this valuation structure, "exceeding expectations" has become the minimum requirement. Investors want "significantly exceeding expectations"—higher AI business growth rates, stronger guidance, and more aggressive long-term goals.

When the market's expectations for a company are raised to the ceiling, even a "good" earnings report can be punished for being "not good enough."

3. New Standards for AI Earnings Reports: Smart Spending, Stable Revenue, Extremely High Guidance

The after-hours movements of SpaceX and AMD reveal a market paradigm shift that is taking place.

AI remains the hottest trading narrative in the U.S. stock market, but the threshold for "participating in AI" has significantly increased. In the past, as long as a company mentioned AI in its earnings report and demonstrated capex growth, its stock price would rise. Now, the market demands a more complete logic:

• No mindless cash burning—capital expenditures must have a clear return path and cannot burn twice the revenue like SpaceX.

• Core business revenue must not be affected—AI investments cannot come at the expense of core business profitability.

• Future expectations must be extremely high—guidance, growth rates, and market share targets cannot be low at all.

Only by meeting all three standards can a company receive market rewards. Missing any one of them may lead to a "good news fully priced in becomes bad news" situation.

4. In Conclusion: The U.S. Stock Market is Becoming "Meme-ified": Tools Are More Important Than Predictions

As the standards for judging earnings reports are pushed to such extreme heights, the characteristics of the U.S. stock market are also undergoing subtle changes.

Any good news is easily pushed to extreme highs—because funds are extremely crowded in the AI narrative, the density of directional bets is very high, and once there is truly unexpected news, buying pressure will rush in.

However, even a slight deviation from expectations can lead to violent corrections—also due to crowding, any disappointment will be amplified into panic selling. The after-hours movements of SpaceX and AMD are typical cases in this "high expectations + low tolerance" environment.

This trend resembles the volatility logic of meme coins in the crypto space more and more—not driven by fundamentals, but by emotions and expectations. In this situation, managing risk becomes more important than ever.

Currently, BIT brokerage has launched an options buying feature that can help investors insure their holdings, strictly locking in downside risk, allowing investors to cope with frequent large fluctuations with ease.

【Risk Warning】The analysis and opinions contained in this article are merely a compilation of market information and personal observations and do not constitute any investment advice, offer, or invitation to offer, nor do they constitute a recommendation for the trading of related securities. The financial data and forecasts cited in the text are as of the publication date; any discrepancies should be based on the company's official earnings reports and publicly disclosed documents. Options trading involves significant risks and is not suitable for all investors; purchased options may expire worthless, resulting in a total loss of premiums. Specific risks should be referred to in the options trading terms. Investors should make their own decisions based on their financial situation and risk tolerance and consult independent professional advisors.

This article is contributed by an external guest author, and the views expressed in it are solely those of the author and do not represent the official position of BIT. BIT makes no guarantees regarding the accuracy or completeness of the content.