Invezz

Why Nvidia stock is surging over 3% today

Why Nvidia stock is surging over 3% today
Utkarsh Roshan
22 Jul 2026, 18:40 PM

powered by

Invezz
NVDA buy

Buy Nvidia (NVDA). The stock is rising on renewed AI-spend expectations ahead of Alphabet and the broader mega-cap capex read-through. Bank of America’s thesis is that Nvidia’s Vera CPU expands its “AI infrastructure” footprint and shifts demand from standalone GPUs to a co-designed system (Vera + Rubin + NVLink + networking). If Alphabet/Microsoft/Meta/Amazon signal steady AI infrastructure budgets and no pricing collapse, NVDA should keep outperforming AMD/Broadcom.

Key Risk: AI capex guidance disappoints or buyers shift to cheaper custom chips that reduce Nvidia’s share of total compute spend.

AMD sell

Sell AMD (AMD). The article flags Vera as Nvidia’s direct push into agentic AI workload measurement/monetization, where Nvidia’s architecture is optimized for time-to-complete. If customers adopt Nvidia’s integrated platform approach, AMD’s CPU opportunity for AI workloads weakens versus Nvidia’s system-level pitch.

Key Risk: AMD wins meaningful AI CPU design wins or customers keep prioritizing AMD’s performance/price for agentic workloads despite Nvidia’s Vera launch.

  • Nvidia shares rose ahead of Alphabet's closely watched quarterly earnings report.
  • Investors await AI infrastructure spending guidance from major technology companies.
  • Bank of America reiterated Buy rating, highlighting Nvidia's integrated AI platform.

Nvidia NVDA shares rose over 3% on Wednesday as investors looked ahead to Alphabet's quarterly earnings for fresh insight into artificial intelligence spending, a key driver of demand for the chipmaker's processors.

The gains followed a 2% rally on Tuesday that lifted the broader semiconductor sector.

Advanced Micro Devices and Broadcom also traded about 2% higher on Wednesday.

The broader US market was little changed as rising oil prices tempered investor sentiment.

The S&P 500 edged up 0.1%, the Nasdaq Composite slipped 0.1%, and the Dow Jones Industrial Average gained 152 points, or 0.3%.

Although Nvidia shares have gained nearly 25% over the past year, the stock has underperformed several other semiconductor companies as investors weigh concerns over growing competition and the sustainability of elevated AI chip pricing.

The earnings season is expected to provide a clearer picture of whether major technology companies plan to maintain their current pace of investment in AI infrastructure.

Alphabet earnings in focus

Investor attention has shifted to earnings from major technology companies, beginning with Alphabet's results after Wednesday's market close.

The report is expected to provide additional clarity on artificial intelligence investment plans that could influence demand for Nvidia's chips.

Companies including Microsoft, Meta Platforms, and Amazon are scheduled to report quarterly results this month, with investors closely monitoring capital expenditure guidance as an indicator of future AI infrastructure spending.

Beyond overall spending levels, investors are also expected to scrutinize commentary on the mix of AI hardware purchases.

Large technology companies have increasingly explored custom-designed processors, often developed with partners such as Broadcom, for certain workloads.

While those chips may not match the performance of Nvidia's graphics processing units, they could reduce dependence on third-party suppliers for specific applications.

Bank of America highlights Vera opportunity

Bank of America Research maintained a Buy rating and a $350 price target on Nvidia, arguing that the company's recently introduced Vera central processing unit expands its position in artificial intelligence infrastructure.

The bank estimates the server CPU total addressable market could reach $170 billion by 2030, roughly four times current levels.

According to analyst Vivek Arya, the launch of Vera marks the beginning of a direct competition with AMD over how agentic AI workloads should be measured and monetized.

"The key question for investors is whether agentic AI is primarily constrained by time-to-complete an agent or number-of-agents-per-rack," Arya wrote.

Nvidia's Vera architecture is designed around the former approach.

The processor combines 88 custom Olympus Arm-based cores with 1.2 terabytes per second of memory bandwidth and 3.4 terabytes per second of on-die fabric bandwidth.

Vera is intended to operate as part of an integrated AI platform that includes Nvidia's Rubin graphics processing unit, NVLink interconnect, Spectrum networking switches, and BlueField networking and storage interface cards.

Bank of America said its bullish view is based on Nvidia's ability to offer a co-designed AI system rather than a standalone processor.