A piece of news about computing power from Meta has punctured the unilateral narrative of momentum trading in AI hardware.

Mars Finance News, July 2 – Meta's stock surged during Wednesday's U.S. stock trading session, putting AI hardware trades through an unexpected stress test. The market had expected a relatively quiet start to July. The second quarter saw strong U.S. stock performance, with the S&P 500 just posting one of its best quarters since the pandemic rebound in 2020. However, just before the U.S. market opened, news that Meta might release or sell "excess computing power" suddenly shifted the market narrative.

This news was positive for Meta itself. The market interpreted it as the company shifting from continuously increasing capital expenditure to emphasizing financial discipline and free cash flow. Meta's stock surged accordingly, with the article noting a single-day gain of around 10%, one of its best single-day performances this year.

But for the AI hardware supply chain, it was a different story. Over the past few months, one of the market's most crowded trades was betting on cloud providers continuously expanding computing power, storage, and data center capital expenditure. If Meta starts releasing excess computing power, investors will naturally ask: Is the demand for AI computing power really as infinitely tight as previously expected? Will cloud providers' capital expenditure still only see upward revisions, with no downward revisions?

UBS trader Christina Dwyer said the Meta event pushed the market narrative toward "stronger financial discipline," while easing concerns about rising capital expenditure. This benefits platform technology companies whose valuations are already near low levels, but weakens the "long-term computing power shortage" logic that had previously supported neocloud, semiconductor, storage, and AI supply chain stocks.

The market reacted quickly. The BofA Neocloud Basket declined notably, with storage stocks and momentum stocks taking a hit. Previously soaring names like SanDisk and Micron were sold off. Micron is particularly viewed as a key indicator: it has held its 20-day moving average since April; if it breaks below, it could open up room for a retreat to the 50-day moving average, implying potential downside risk of about 20%.

This adjustment quickly turned into a momentum trade cleanout. BTIG's Jonathan Krinsky noted that the Bloomberg Mag7 index relative to the Philadelphia Semiconductor Index (SOX) saw its largest single-day rebound since 2015. In other words, capital flowed out of chips, storage, and high-beta AI hardware stocks and back into large-cap platform technology stocks. Goldman Sachs' high-beta momentum basket fell about 9% in one day, and the long-short high-beta momentum combination fell about 10%, approaching its worst performance since the vaccine announcement shock in 2020.

The Meta event reminds investors that cloud providers, who bear the real capital expenditure, also calculate returns. Once the certainty of upward capex revisions weakens, the segments of the AI hardware chain with the largest gains and most crowded valuations will come under pressure first.

However, capital hasn't completely left the AI theme. Software stocks have strengthened relative to semiconductors, and Bitcoin has also rebounded due to capital outflow from AI/storage momentum trades. Another beneficiary is capacitor and other AI bottleneck assets, indicating that the market is still looking for scarce links in computing infrastructure, just no longer indiscriminately chasing storage and chip momentum stocks.

Another risk to note in the current market is the very low liquidity. Goldman Sachs' trading desk said that in June, the top-of-book liquidity for S&P E-mini futures fell 33% month-over-month, while U.S. stock trading volume hit its highest since 2026. This means the market appears active in volume, but actual capacity to absorb trades is weaker; once large sell orders appear, prices are more prone to sharp fluctuations.

The real meaning of the Meta event may not be about Meta itself, but that it hit the most sensitive spot in the AI market narrative: whether capital expenditure will continue to grow without limit. AI demand hasn't disappeared, but the market is starting to distinguish between two types of companies: those that can steadily monetize from computing investment (platform companies), and those that have already fully priced in computing shortage expectations (hardware and storage suppliers).