Nvidia Forecasts Revenue Above Expectations, Announces $80 Billion Stock Buyback Plan
On May 20, U.S. AI chip giant Nvidia projected its second-quarter revenue would surpass Wall Street expectations and announced an $80 billion stock buyback program. Despite the strong performance, its stock dipped slightly in after-hours trading. Analysts note that while Nvidia continues to lead the AI chip market, it faces fierce competition from tech giants like Google, Amazon, and AMD. These companies are actively developing their own chips, especially for the larger 'inference' market, posing a challenge to Nvidia's long-term dominance. Nvidia is moving to solidify its position by launching new products and ensuring supply chain stability.
📋 Article Processing Timeline
- 📰 Published: May 21, 2026 at 09:52
- 🔍 Collected: May 21, 2026 at 10:01 (9 min after Published)
- 🤖 AI Analyzed: May 21, 2026 at 10:13 (11 min after Collected)
American artificial intelligence (AI) chip giant Nvidia today forecasted second-quarter revenue that will exceed Wall Street expectations and announced an $80 billion stock buyback program. Nvidia's stock fell slightly by 0.2% in after-hours trading today.
According to Reuters, based on data compiled by the London Stock Exchange Group (LSEG), the world's most valuable company expects second-quarter revenue to reach $91 billion, plus or minus 2%, higher than the market estimate of $86.84 billion.
Nvidia's financial reports are widely seen as an indicator of the AI market's health, as its chips are used in nearly all major data centers worldwide, driving the largest and most advanced AI models.
Jacob Bourne, an analyst at market research firm eMarketer, said, "Nvidia has once again delivered a stellar earnings report, but at this point, the market has already priced in its repeated outperformance. The lingering question is whether Nvidia can convince investors that the AI build-out boom can extend into 2027 and 2028, especially as the market narrative shifts to inference workloads and as chips from rivals like Google, Amazon, AMD, and Intel enter the competition."
Nvidia also stated that its quarterly cash dividend will be increased from 1 cent per share to 25 cents per share.
Spending on AI infrastructure continues to grow rapidly. U.S. tech giants including Alphabet, Amazon, and Microsoft are expected to spend over $700 billion on AI this year, a significant increase from about $400 billion in 2025.
While these companies heavily rely on Nvidia's expensive processors, they are also investing funds in developing their own custom chips to run models, posing a risk to Nvidia's long-term dominance in the chip industry.
Those chips are primarily used for inference, the process where an AI answers a user's query, a market segment much larger than training.
Nvidia faces competition not just from tech giants but also from chip competitors like Intel and Advanced Micro Devices (AMD), who have joined the fray and claim that the inference market presents a massive revenue opportunity.
Headquartered in Santa Clara, California, Nvidia has taken steps to defend its position, unveiling a new central processing unit and AI system in March that utilizes technology from Groq, a chip startup specializing in inference.
During a quarterly earnings call with financial analysts, Nvidia CFO Colette Kress stated that the market for Nvidia's central processing units (CPUs) is about $200 billion, and the company "expects to achieve nearly $20 billion in total CPU revenue" this fiscal year.
Nvidia is also investing heavily to ensure it does not face supply chain bottlenecks amid a global memory chip shortage. Nvidia said today that its supply size increased to $119 billion in the first fiscal quarter, up from $95.2 billion in the previous quarter.
According to data compiled by LSEG, Nvidia's first-quarter revenue reached $81.62 billion, exceeding the average analyst estimate of $78.86 billion.
First-quarter data center revenue was $75.2 billion, higher than the average analyst estimate of $72.8 billion. Adjusted earnings per share were $1.87, above the market expectation of $1.76.
Nvidia also reported $30 billion worth of cloud computing agreements, up from $27 billion in the previous quarter. Nvidia said these agreements are to support research and development efforts. In a research note last year, Seaport analyst Jay Goldberg pointed out that such commitments could represent a 'backup' plan, where Nvidia agrees to pay cloud computing companies that buy its hardware to access the excess computing resources generated when they run Nvidia's systems.
According to Reuters, based on data compiled by the London Stock Exchange Group (LSEG), the world's most valuable company expects second-quarter revenue to reach $91 billion, plus or minus 2%, higher than the market estimate of $86.84 billion.
Nvidia's financial reports are widely seen as an indicator of the AI market's health, as its chips are used in nearly all major data centers worldwide, driving the largest and most advanced AI models.
Jacob Bourne, an analyst at market research firm eMarketer, said, "Nvidia has once again delivered a stellar earnings report, but at this point, the market has already priced in its repeated outperformance. The lingering question is whether Nvidia can convince investors that the AI build-out boom can extend into 2027 and 2028, especially as the market narrative shifts to inference workloads and as chips from rivals like Google, Amazon, AMD, and Intel enter the competition."
Nvidia also stated that its quarterly cash dividend will be increased from 1 cent per share to 25 cents per share.
Spending on AI infrastructure continues to grow rapidly. U.S. tech giants including Alphabet, Amazon, and Microsoft are expected to spend over $700 billion on AI this year, a significant increase from about $400 billion in 2025.
While these companies heavily rely on Nvidia's expensive processors, they are also investing funds in developing their own custom chips to run models, posing a risk to Nvidia's long-term dominance in the chip industry.
Those chips are primarily used for inference, the process where an AI answers a user's query, a market segment much larger than training.
Nvidia faces competition not just from tech giants but also from chip competitors like Intel and Advanced Micro Devices (AMD), who have joined the fray and claim that the inference market presents a massive revenue opportunity.
Headquartered in Santa Clara, California, Nvidia has taken steps to defend its position, unveiling a new central processing unit and AI system in March that utilizes technology from Groq, a chip startup specializing in inference.
During a quarterly earnings call with financial analysts, Nvidia CFO Colette Kress stated that the market for Nvidia's central processing units (CPUs) is about $200 billion, and the company "expects to achieve nearly $20 billion in total CPU revenue" this fiscal year.
Nvidia is also investing heavily to ensure it does not face supply chain bottlenecks amid a global memory chip shortage. Nvidia said today that its supply size increased to $119 billion in the first fiscal quarter, up from $95.2 billion in the previous quarter.
According to data compiled by LSEG, Nvidia's first-quarter revenue reached $81.62 billion, exceeding the average analyst estimate of $78.86 billion.
First-quarter data center revenue was $75.2 billion, higher than the average analyst estimate of $72.8 billion. Adjusted earnings per share were $1.87, above the market expectation of $1.76.
Nvidia also reported $30 billion worth of cloud computing agreements, up from $27 billion in the previous quarter. Nvidia said these agreements are to support research and development efforts. In a research note last year, Seaport analyst Jay Goldberg pointed out that such commitments could represent a 'backup' plan, where Nvidia agrees to pay cloud computing companies that buy its hardware to access the excess computing resources generated when they run Nvidia's systems.