Behind the sharp rise of NVIDIA: Morgan Stanley recommends buying the dip, Saudi Arabia and the United Arab Emirates compete for GPUs.

Overnight, the stock price of NVIDIA had a wild celebration in the US stock market, rising over 7% at the close and continuing to rise by 1.8% after hours, currently trading at $445.57 per share.

There are two major positive factors behind the surge in NVIDIA's stock price: endorsement from Morgan Stanley and competition among Middle Eastern tycoons for NVIDIA graphics cards.

Morgan Stanley: NVIDIA remains the top choice with significant upside potential in the future

On Monday local time, Morgan Stanley analysts released a report reiterating that NVIDIA is still the "top choice" company in their second-quarter earnings report. Morgan Stanley's report provided better-than-expected forecasts.

Due to supply constraints and ongoing market concerns about the economy, NVIDIA's stock price has fallen more than 10% this month. However, Morgan Stanley analysts predict that NVIDIA will still benefit in the long run.

In their report, Morgan Stanley analysts wrote:

We believe the recent sell-off presents a good buying opportunity because despite supply constraints, we still expect significant growth in this quarter's performance. More importantly, the outlook for the next 3-4 quarters is strong.

Against the backdrop of a massive shift in spending towards artificial intelligence, NVIDIA remains our top choice, and the unique supply-demand imbalance will continue to exist in the coming quarters.

NVIDIA is currently valued at over $1 trillion. After experiencing a market downturn in 2022, NVIDIA's stock price has skyrocketed, surging more than 200% year-to-date, outperforming most technology stocks.

Analysts also stated:

Most importantly, this is a very positive situation, as October's data is entirely dependent on supply, and the upper limit of buyer consensus has also been controlled.

We believe the numbers will increase to a level where this stock trades at a price-to-earnings ratio similar to the semiconductor industry's upper limit, leaving significant room for further upside in the future.

NVIDIA's stock price has doubled this year, and the company will announce its second-quarter earnings report on August 23.

Middle Eastern tycoons fiercely compete for NVIDIA graphics cards

According to the Financial Times on Tuesday, Saudi Arabia and the United Arab Emirates are competing to purchase thousands of high-performance NVIDIA graphics cards, joining the global artificial intelligence arms race.

The report cited informed sources stating that Saudi Arabia has purchased at least 3,000 NVIDIA H100 graphics cards through the public research institution King Abdullah University of Science and Technology (KAUST), with each card priced at $40,000.

Meanwhile, the United Arab Emirates has also ensured that it will obtain thousands of NVIDIA graphics cards and has developed its own open-source large-scale language model called Falcon at the Masdar City-owned Institute of Technology Innovation in Abu Dhabi. According to multiple sources close to Nvidia and its manufacturer TSMC, Nvidia is expected to ship approximately 550,000 units of the H100 graphics card worldwide in 2023, with the main market being US tech companies.

According to two individuals familiar with the university's AI lab, Kaust is set to receive 3,000 of these specialized graphics cards by the end of 2023, with a total value of around $120 million.

Kaust also possesses at least 200 Nvidia A100 graphics cards and is currently building a supercomputer called Shaheen III, which is expected to be operational later this year. The computer will run on 700 Nvidia super graphics cards, specifically the GraceHoppers model.

Furthermore, several sources closely related to the university revealed that Kaust plans to utilize these graphics cards to develop its own large-scale language model, similar to OpenAI's GPT-4.

Meanwhile, the United Arab Emirates became the first country to establish a dedicated AI department in 2017 and launched the "Generative AI Guidelines" to enhance its global position as a pioneer in technology and AI.

The Falcon model is now available online for free and underwent over two months of training on 384 A100 graphics cards earlier this year.