In the past week, the market’s enthusiasm for artificial-intelligence hardware has been dampened by a rising awareness of financing hurdles. Baird’s technology strategist Ted Mortonson warned that investors are tiring of the “bank of Nvidia, the bank of AMD and the bank of Broadcom,” hinting that the cost of capital is now a decisive factor for AI-related equities.
The heightened scrutiny follows an anthropic research meeting where the AI startup Anthropic disclosed a third-quarter revenue surge to $16.8 billion, yet the broader narrative has shifted from revenue growth to the sustainability of cash flows under tighter credit conditions.
Financing pressures reshape AI-chip rally
Mortonson highlighted that companies such as CoreWeave, Nebius and Oracle are approaching the lower bound of free-cash-flow generation. Their heavily leveraged balance sheets mean that a higher cost of capital translates directly into stock pressure. The same dynamic is evident in the recent tumble of CoreWeave shares, which fell more than 15 % after concerns were voiced about its debt load and the ability to meet rising interest expenses.
Even chipmakers that posted stellar earnings this quarter – Samsung, Micron and ASML – saw their shares slide as portfolio managers trimmed positions that, despite “blowout numbers,” now carry a heavier financing burden. Gartner’s forecast that many neocloud providers could cease independent operations by 2030 added another layer of doubt, reinforcing the view that investors are reallocating capital away from capital-intensive AI infrastructure.
Market reaction on October 16, 2026
On the trading day of October 16, 2026, the Dow Jones Industrial Average climbed 1.1 %, the S&P 500 added 0.3 %, and the Nasdaq Composite managed a modest gain after previously slipping more than 1 %. The rally was anchored by a drop in the 10-year Treasury yield to 5.09 %, signaling a flight to safety that eased the debt-financing angst surrounding AI stocks.
CoreWeave managed to claw back part of its losses after Nvidia CEO Jensen Huang publicly expressed optimism about the data-center operator, noting that its cost of capital was improving. Nevertheless, the broader sector of AI-infrastructure providers, including Vertiv, Comfort Systems and Quanta Services, continued to post losses, collectively shedding roughly $94 billion since early October.
Choosing the long-term AI winner: Nvidia or Broadcom?
Two titans dominate the AI-hardware conversation: Nvidia and Broadcom. Nvidia’s fiscal 2027 second-quarter revenue jumped 106 % to $96.2 billion, with its data-center segment alone delivering $89 billion, up 117 %. The company’s integrated platform – GPUs, networking gear, the CUDA software suite and the upcoming Vera Rubin system – underpins a competitive moat that extends beyond raw processing power.
Broadcom, meanwhile, has built a lucrative niche around custom ASICs and networking chips, reporting $29.6 billion in revenue for fiscal 2026 Q3, an 86 % year-over-year increase, and generating $13.7 billion in free cash flow. While its processors lack the breadth of Nvidia’s GPUs, they offer efficiency and cost advantages that appeal to hyperscalers looking to diversify their supply chain.
Analysts suggest that the higher-cost financing environment may favor Broadcom’s lower-capex, ASIC-centric model, yet Nvidia’s expansive ecosystem and cash-rich balance sheet give it a resilient edge. For investors seeking a five-year horizon, the debate hinges on whether the market rewards breadth and integration (Nvidia) or specialization and cost efficiency (Broadcom).
