Summary
Which of the market dominant and so -called “beautiful 7” shares are perhaps the best investments? During the first phase of unbridled AI enthusiasm all were allowed to be loved because their chance seemed to be almost unlimited. The market is now more skeptical about observed AI winners, which has led to a natural tendency to rank the group on their prospects. We see Amazon going up in the ranks, while Nvidia continues to control the top position. All may 7 names have something in common. They are all dominant in at least one core competence; And they have used the immense cash flows that are thrown away by their core competencies to invest and develop the generative AI opportunity. “Ranking” of the MAG 7 on their prospects in the medium term therefore includes assessing and cherishing the ability of each company to defend and cherish its core activities while using that power to win shares within the AI ecosystem. For the list of this week we analyze and arrange the MAG 7 (from top to bottom) on their apparent successive prospects. We expect that our rankings for the MAG 7 are fluent, but did not provide any movement at the first or last place.* NVIDIA (NVDA): We will continue to place Nvidia at the top of the MAG 7 list. Unlike the other six names, the core competence of Nvidia is artificial intelligence. The need for realistic representation in GPU-based gaming, the founding company of NVIDIA, laid the foundation for the Applications acceleration and the mass parallel computing that are cornerstones of training large language models (LLMS) and making secludes possible. We rank NVIDIA number one based on the evidence in the company’s own growth numbers, which are fantastic. In his recent fiscal 4Q25, $ 39 billion sales grew annually by 78%. During his GTC event in March 2025, NVIDIA predicts that the income from Data-Center will amount to $ 1 trillion annually in the coming years, and that the entire global industrial infrastructure was a chance of $ 50 trillion for AI extension. NVIDIA also offers much more than its leading GPUs, including the CUDA software library, and is “turbo chargers” agentic AI development with open-racing models, platforms and partnerships. The best proof of Nvidia’s momentum is the growth of the business infrastructure companies that use NVIDIA solutions to support the mainstreaming of AI. The income from the Micron data center tripled year after year and now exceeds 50% of the total business income. Micron raised the tame for his occasion with high bandwidth (HBM) to $ 35 billion from March 2025, an increase of $ 20 billion about half a year ago. Under the leading American GPU server providers, Dell Technologies booked $ 10 billion in AI Seze Turnover in FY25 and expected $ 15 billion in FY26. Hewlett-Packard Enterprise left Fiscal 1Q25 with $ 8.3 billion in cumulative AI systems and service assignments. The leaders in Datacenter Interconnects (DCIS), Broadcom and Marvell both report explosive growth in these categories. * Meta platforms (Meta): Meta platforms in second place because it is one of the most successful in the use of its core competence (social media) in a leading role in building LLMs, including multimodal models. Snap has a limited chance and X is impeded by self -inflicted wounds and observed muskeberry. Only companies such as Bytedance (Tiktok) based in Asia pose a real threat. We see some risk for the core activities of apparently endless EU and American congress -; And meta’s social media cornerstones (Facebook, Instagram and WhatsApp) can be carved neatly in three. However, we don’t see that happening. Meta has been successfully played from the expensive obsession of the Zuckerberg with the metaverse to the more-practical and rapidly developing AI opportunity. Unlike AWS, Microsoft Azure and Google Cloud, Meta is successfully switching on and promoting leading LLMS, even if it is not a large provider of hybrid cloud services. We see Meta primarily using AI internally to improve operational efficiency and personal targeting to optimize the user base of the company’s social media, which is greater than 3.3 billion. Facebook Reality Labs, with the Gen AI activities of Meta, achieved a turnover of $ 2.1 billion in 2024, only 1% of the total meta income. Currently this is mainly a hardware-based company (Quest VR headsets, Ray-Ban AR glasses), but we see growth opportunities in the long term as Llams LLMS from Meta Go Mainstream. As Apple has capitalized on the huge iOS-installed base and focused the individual with iCloud, Meta could offer a personalized AI offer to its more than three billion users.* Amazon.com (AMZN): Amazon is in third place. The online retail activities of Amazon and in particular the most important activities are without a global pear. The combined work margin (America and International) for 4Q24 was 6.1%, by far the highest in our model that went back to the establishment of the company. The media of Prime now compete in Netflix in quality and volume. AWS is the leading CSP worldwide, and the rising margins and cash flows finance the opportunity of the company. Amazon was quickly replaced the head of AWS in May 2024 when the board noticed that the company is lagging behind in Gen AI. AWS has an annual turnover of $ 110 billion and the margins are at or almost record highs. AI has produced a cut market both internally and in collaboration with Anthropic and others, where customers can choose from more than 100 LLMs; Trainium and Trainium 2 gears; and Amazon Nova, an extensive family of fundamental models. * Apple (AAPL): Fourth place goes to Apple. Apple’s core competence is leadership in technology devices and services worldwide. Apple has been somewhat disappointing in his early efforts in AI. The company even goes beyond its usual dependence on internal possibilities to enable support from companies such as OpenAI. But we have long regarded Apple as a product perfector, not a product pioneer; Early iteration mobile phones such as Dynatac and Startac are surprised, but they have long since disappeared while the iPhone is growing in dominance. Investors are also concerned about the falling Chinese sale of Apple. But in the most recent tax 1q25, Apple income grew in all other regions. The huge iOS basis of the company is more than 2.2 billion, and offers a fertile soil for continuous growth in services.* Microsoft (MSFT): Microsoft is in fifth place. Microsoft has used the core competence in Enterprise software in Azure, the number two CSP behind AWS. Like all the companies mentioned above, Microsoft is challenged to get enough stock of advanced Blackwell products to build and run the most advanced models. The cloud activities of Microsoft have never been as profitable as AWS, although that can be partly that assets are assigned to the Cloud Division. Microsoft has had a few missteps when rolling out his AI models, and bone
