The historical ascent of the Nvidia stock has left no room for the error.
For more than two years, no print has dominated Wall Street more than those who have treated the rise in artificial intelligence (AI). The empowerment of software and systems with the capacity to reason and act alone gives this technology a long -term ceiling apparently unlimited.
While the adoption rate and the usefulness of AI solutions remain fluid, PwC analysts see this technology which completely changes the world growth trajectory by 2030. Come the turning point of the decade, the PWC Prize The report estimates a positive impact of 15.7 billions of dollars on the gross domestic product, all thanks to the AI.
Image source: Getty Images.
A tart of $ 15.7 billion is large enough to accommodate a long list of winners. However, no company was a more direct beneficiary of the AI revolution than the giant of the graphic processing unit (GPU) Nvidia (NVDA -1.51%)). Nvidia went from a company of $ 360 billion at the end of 2022 to a company that came to a stone throwing a market capitalization of 4 dollars on an intra-day basis just over two years later.
Although it was only denier Nvidia was a success of AI, there is a real possibility that leading companies could soon become a victim of its own success.
The ascent of Nvidia is nothing less than the manual
Before connecting the points of the way in which Nvidia’s perfect operating expansion can potentially take place on itself, it is important to understand the context of the way in which it has become the foundation of the AI revolution.
At the heart of Nvidia’s success is his Hopper (H100) and the new generation Blackwell GPUs. The first has indeed become the standard in data centers accelerated by AI in the previous two years. Meanwhile, the recently deployed Blackwell GPU architecture significantly improves calculation capacities for generative AI solutions and quantum IT, while being considerably more economical than its predecessor chip.
To rely on this point, no other flea manufacturer has particularly approached, to challenge the versatile functionality of Nvidia equipment. While others could offer close competitive advantages in certain tasks, no other GPU has really disputed the hopper or Blackwell on the basis of calculation performance in the AID data centers.
Adding fuel to fire, Nvidia was perfectly positioned to benefit from the rarity of A-GPU. Despite the best efforts of the world cutting -up manufacturing company Manufacture of Taiwan semiconductors To quickly extend its chip capacity on the sub -subsratt (Cowos) – Cowos is essential to pack the main bandwidth memory required for high -company data centers – A -GPU demand has constantly overwhelmed their supply. Consequently, Nvidia ordered a price premium of 100% (or more) for its AI GPU, compared to its competition.
The Cuda de Nvidia software platform links things to a beautiful arc. This toolbox, used by developers to maximize the performance of their GPUs, as well as to build major language models, was essential to keep customers within the NVIDIA products and services ecosystem.
Nvidia can be a victim of its own success
Although the overwhelming The consensus among Wall Street analysts is that this darling of artificial intelligence is higher – the target of average analyst price from 171 points to 50% – two viable paths exist when Nvidia becomes a victim of its own success, and its course in action suffers.
To start, NVIDIA GPUs are considered to be superior, in terms of IT capacities. While short-term demand continues to overwhelm the supply for A-GPU, Hopper and Blackwell’s computer superiority could considerably slow down any future upgrading / replacement cycle.
Think about it this way: when Apple He first went out with the iPhone, consumers and businesses jumped at the opportunity to move on to this revolutionary apparatus. Over time, the new generation versions of the iPhone have presented upgrades, but they were much less dramatic or impressive, compared to what the original iPhone did for consumers and businesses in 2007. With Nvidia equipment being more expensive and faster than its peers, the impulse for companies to upgrade their central AI-Data equipment can be pushed for years. This is not good news for a business that Wall Street expects to support a breathtaking growth rate.
Image source: Getty Images.
The second potential problem for Nvidia is that Wall Street could punish it for simply giving up part of its monopoly market share in AI-Data centers in the next quarters.
On the basis of an analysis of the SEMI-CONICUTER SEMI-CONNEMBER SOCIETY, NVIDIA represented 98% of GPUs shipped to corporate data centers in 2022 And 2023. Even with tastes Advanced micro-apparents By increasing the production of its AI-Accelerating Chips instinct series, Nvidia is always likely to have ordered an almost monopoly share of the “brains” of AA data centers last year.
The problem for NVIDIA is that many of its best customers by net sales, which mainly include members of the “Magnifiment Seven”, develop their own A-GPU and IA solutions which can be used in their data centers. Even if these internal developed chips are not a threat to Hopper or Blackwell’s computer speed, and that it does not plan to sell outside, they could easily make Nvidia lose on the real estate of the future data center. These chips are both cheaper and more easily available than Nvidia equipment.
If future upgrade cycles become more written and / or that Wall Street’s Darling loses a precious real estate from the data center to its own best customers, what is undoubtedly its greatest competitive advantage – rarity of IA -GPU – will disappear. This means less in the manner of pricing power and lower future margins.
Nvidia’s historical ascent has left no room for error. If the growth rate of the company was to slow down, or even if its gross margin continues to go from a peak of 78.4%, it could quickly become a victim of its own success.
Sean Williams has no position in the actions mentioned. The Motley Fool has positions and recommends micro advanced devices, apples, nvidia and the manufacture of Taiwan semiconductors. The Motley Fool has a policy of disclosure.
