CoreWeave just reignited the AI trade — can the NAS100 rally survive the spending backlash?

CoreWeave has given AI investors something they badly needed: evidence that huge infrastructure spending is still translating into rapid revenue growth. The AI cloud provider reported third-quarter revenue of $3.66 billion, up 168% from a year earlier, while revenue backlog reached $132.5 billion and full-year guidance was raised to $13.0–13.3 billion.

The result helped revive sentiment across AI-linked stocks after weeks of selling pressure. Nvidia, Broadcom, Micron and other technology names rebounded as investors took CoreWeave’s numbers as confirmation that demand for AI computing capacity remains strong.

The rebound does not remove the concerns that pushed NAS100 lower in the first place. CoreWeave still expects $37–39 billion of 2026 capital expenditure, while interest expense alone is expected to exceed $1 billion in the fourth quarter. The same tension is visible across Big Tech, where companies are still spending aggressively on chips, cloud infrastructure and data centres while investors become increasingly focused on cash flow and returns.

For NAS100 traders, the argument has therefore moved beyond whether AI demand exists. The more important question is whether companies can convert that demand into enough profit and cash flow to justify the scale of the spending required.

Recent weakness across AI stocks was never primarily about customers abandoning artificial intelligence. The concern was that companies were committing increasingly large amounts of capital without showing clearly enough how quickly that investment would generate acceptable returns.

CoreWeave’s latest quarter helped answer part of that concern because revenue rose by more than $1 billion from the previous quarter, while adjusted operating income reached $379 million and the operating margin moved into double digits sooner than expected. Revenue backlog of $132.5 billion also showed that customers continue to commit substantial spending even before all of CoreWeave’s new capacity comes online.

The company also expects around $9 billion of customer prepayments and deposits over the next three months, which gives it another source of funding as it expands data-centre capacity. Those commitments suggest that cloud customers are still willing to lock in future computing resources rather than pulling back after the recent AI spending backlash.

CoreWeave itself does not dominate NAS100, but the companies supplying and benefiting from its expansion do. New data-centre capacity requires more GPUs, networking equipment, memory and storage, which creates direct demand across Nvidia, Broadcom, Micron and other semiconductor names.

CoreWeave added more than 200MW of active power during the quarter, taking its operating footprint above 1.7GW, while contracted power climbed above 4.6GW. That expansion supports the idea that AI infrastructure spending is still broadening rather than stalling.

For Nvidia, the connection is especially direct because new cloud capacity requires large volumes of advanced accelerators. Broadcom benefits from networking and custom AI hardware, while memory suppliers gain as larger AI clusters require more high-performance storage and memory. CoreWeave’s results therefore provided a positive read-through for the wider AI ecosystem that carries much more weight inside NAS100.

The same results that reassured investors about demand also showed why concerns about AI economics have not disappeared. CoreWeave expects $37–39 billion of capital expenditure this year, several times its expected annual revenue, while interest expense continues rising as the company funds expansion through debt, convertibles and equity.

That model can work if new capacity becomes revenue-generating quickly and customer demand stays strong. Delays, lower utilisation or higher financing costs could make the economics much less attractive, particularly because large amounts of capital need to be committed before the associated revenue arrives.

Big Tech faces the same issue from a stronger financial position. Meta, Microsoft, Alphabet and Amazon are all investing enormous sums in AI infrastructure, but investors are now paying much closer attention to whether cloud growth, advertising gains and free cash flow are keeping pace with those budgets.

CoreWeave improves confidence in the demand side of the equation, but it does not settle the question of how efficiently the broader industry can turn infrastructure spending into durable returns.

Nvidia still provides one of the cleanest links between CoreWeave’s growth and the index because every expansion in AI cloud capacity increases demand for advanced computing hardware. CoreWeave’s backlog, power commitments and spending plans all point towards continued infrastructure buildout, which supports expectations for further GPU demand.

The recent sell-off in Nvidia reflected concern that hyperscaler capex could eventually slow after several years of exceptional growth. CoreWeave’s numbers push back against that fear by showing that AI cloud customers are still committing heavily to future capacity.

The next test is whether Nvidia’s own earnings confirm the same picture. Strong data-centre revenue and guidance would reinforce the idea that the recent NAS100 pullback was driven by valuation concerns rather than a deterioration in demand, while weaker guidance could quickly revive fears that the AI spending cycle is starting to slow.

The latest rebound does not erase the damage from the recent technology sell-off. NAS100 remains below its October highs after investors reduced exposure to AI-heavy growth stocks and became more cautious about the amount of capital being committed to infrastructure.

The index now needs more than strong revenue growth to regain momentum. Microsoft must continue showing that Azure growth is strong enough to justify rising data-centre investment, Amazon needs AWS expansion to offset pressure on free cash flow, Meta has to demonstrate that AI spending is improving advertising and engagement, while Nvidia needs customers to keep buying enough hardware to support high expectations.

If those companies continue delivering strong revenue, margins and cash generation, CoreWeave’s quarter could look like the first sign that the recent correction went too far. If spending keeps rising faster than returns, investors may continue treating AI infrastructure as a source of risk rather than an automatic growth catalyst.

The AI rebound is also taking place while US rates remain restrictive, which increases the pressure on expensive growth stocks and companies relying heavily on external financing. CoreWeave’s rising interest expense shows how quickly higher borrowing costs can become part of the earnings story when expansion depends on debt.

Large NAS100 companies generally have stronger balance sheets, but higher yields still affect valuations because investors can earn more from lower-risk assets while applying a higher discount rate to future profits.

That makes cash generation increasingly important. Companies that can fund AI investment internally and still produce strong free cash flow are likely to receive more support than businesses that need repeated capital raises or years of spending before returns become visible.

• Nvidia earnings: Strong data-centre demand would reinforce CoreWeave’s positive signal.

• Hyperscaler capex: Microsoft, Amazon, Meta and Alphabet need to show that rising investment is still producing revenue growth.

• AI cloud demand: Continued growth from CoreWeave and other infrastructure providers would support the broader spending cycle.

• Free cash flow: Investors are increasingly focused on how much cash remains after infrastructure spending.

• Interest rates: Higher financing costs can pressure both AI economics and growth-stock valuations.

• Data-centre execution: Power access, construction delays and funding costs can still disrupt otherwise strong demand.

Trading NAS100 with Mitrade

CoreWeave’s results have restored some confidence in the AI infrastructure story, but the market still has to decide whether strong demand is enough to justify the scale of current spending.

Eligible Australian clients can use Mitrade CFDs to take a long position on NAS100 if they expect the technology rebound to continue or a short position if concerns around valuations, capex or financing return.

That flexibility is useful while the same AI spending cycle is producing both bullish and bearish catalysts. Strong cloud growth, improving margins or better cash generation could support the index, while rising debt, weaker guidance or disappointing returns on infrastructure investment could push it lower.

Stop-loss and take-profit orders can help define exit levels, while pending orders allow positions to open when selected prices are reached. NAS100 can move sharply around technology earnings, inflation releases and changes in Fed expectations, so leverage and position sizing remain important.

Mitrade is regulated in Australia by ASIC. Index CFDs are leveraged products, so both potential gains and losses can be magnified.

You might be interested in…