AI is overvalued, akin to the internet during the Dot Com boom and bust. But this time the Nasdaq Composite will fall somewhere above 50 % from its peak, contrary to the 80 % + fall of the Nasdaq from its peak in the Dot Com boom and bust.
Artificial Intelligence, AI is just not ready to produce the revenues and profits Wall Street equity research and fixed income research analysts seem to be inputting in their models. Or at least, not as quickly as openly assumed.
AI is simply very difficult to scale. Yes, the picks and shovels, namely the graphical processing units, computer processing units and other computer chips needed can be sold for hundreds of billions of USDs. That is why, NVIDIA's current market capitalization is 2.26 trillion USD currently.
The quantitative models, which are basically the AI and which power the computer chips are very difficult to scale. Nonlinear modelling is extremely difficult and requires huge computing power, which according to Wolfteam Ltd.'s estimates is not yet available.
All in all, AI, like almost every branch of science develops in leaps. The first leap came with Bitcoin and the blockchain algorithm, which huge computing power needs created an explosion in AI computer chips manufacturing. Currently, data centers computing needs provide from a smaller AI leap.
The next AI leap will come in more than 10 years, according to Wolfteam Ltd.'s projections and estimates. Much depends on how Wall Street research analysts and investors model, discount the future in their models and expectations. Namely, much hangs on whether investors are ready to wait long enough, 5-7 years for their AI investment ideas to come to fruition.
Apple
Currently Apple uses AI via the Google search engine. Apple adapts its hardware M1, M2 and M3 latest computer chips, so they can process AI tasks most efficiently. Apple shuns away from gathering user data to drive AI applications. There are news reports that Apple will soon integrate ChatGPT in the App Store. Actually, via the App Store apple has gathered huge amount of information, which could prove helpful in AI development. App Store, namely Apple's services business is very lucrative for Apple with its 30 % fee cut. Actually, App Store or Apple's services business is at the heart of Apple's future strategy. App Store revenue is 27 billion USD a year for Apple with a staggering 80 % net profit margin, according to court documents. So Apple seems to make circa 18 billion USD in profit only from the Apple store each year. There is still a tremendous growth opportunity for App Store revenue, which is part of the services revenue business segment of Apple. And App Store recommendations are driven by artificial intelligence, AI, so Apple's services business is basically an AI business. Apple has consistently shown genius in developing ingenious computer chips. Now that hardware engineering genius is being applied to developing AI chips. M3 Apple chips is exactly that - a chip developed exactly with AI in mind. A risk before Apple is that 80 % of its Macintosh chips, iPhone computer chips, Apple tablet chips are produced in Taiwan. A China Taiwan conflict could seriously disrupt Apple, by cutting its market capitalization with more than 60 % in a month. That is why Apple' AI chips development is so pivotal for Apple's future. Apple has tried to diversify chip production in various countries.
Microsoft
Microsoft is currently leading the AI field with its 49 % stake involvement in ChatGPT. ChatGPT is a chat robot, a quasi search engine which gathers, analyses users' data and data from the internet to produce search query results. Initially ChatGPT was taking the world by storm and Microsoft, being almost a majority, but definitely a controlling shareholder was leading the AI field. ChatGPT's technology uses much of Google's Large Language Models technology and is pulling the AI field forward. Microsoft benefited tremendously from the ChatGPT and AI and is currently the most valuable listed company in the world surpassing Apple's nearly 3 trillion USD market capitalization. Microsoft owns 49 %, non majority stake in ChatGPT and thus skirts the accusations that it could misappropriate users' data by excessively mining it. Microsoft however already stores and analyzes users' data via its Bing search engine. So Microsoft has to juggle AI opportunity with users' data concerns and for now Microsoft is doing that brilliantly. That said, ChatGPT's inertia has stalled lately after the recent management turmoil in ChatGPT. That said, Microsoft is left relatively unscathed from ChatGPT's problems, despite actually Microsoft exercising huge influence over ChatGPT's management decisions. The main driver of Microsoft's revenue and profits is the cloud computing division. It is important to note that the cloud Microsoft Office known as Office365 is housed into the cloud computing division. In order for its second crown jewel Microsoft Office, apart from Microsoft Windows to function well, Microsoft, Inc has to deploy complicated artificial intelligence algorithms, which are based on machine learning and are pivotal for Microsoft Office Cloud version to function smoothly. In addition, Microsoft has its stand alone cloud computing, storage analysis business known as Azure. There the latest artificial intelligence, machine learning algorithms are used to store data, analyze it and produce predictive analytics. And for Azure all this is priced in increments which makes it affordable for small, medium sized and large companies to buy. Microsoft for its part gets via Azure and Microsoft Office cloud significant, even huge amounts of new information on which to train and perfect its AI algorithms.
Alphabet
Alphabet Inc is also leading the AI pack with its Google search engine, which has utilized AI to produce search results for more than a decade now. Google has developed Gemini, its own Large Language Models Chat bot to compete efficiently with ChatGPT in the AI race. Alphabet Inc, actually most probably has the largest data treasure trove of any company in the world, which can be modeled via AI, machine learning algorithms to produce actionable insights, better search results and ultimately higher revenues and profits for Alphabet, Inc. Alphabet makes 13 billion USDs in revenue from the Play Store with a circa 70 % net profit margin according to court documents. All of the recommendations in the Play Store business are driven by AI, so the Play Store is actually an AI business. The Play Store is a tremendous future growth opportunity for Alphabet, Google's parent. Recently Alphabet launched its Bard chat bot and Gemini Large Language Model application. Large Language Models, LLMs were actually developed first mostly at Alphabet's subsidiary Google, the LLM process lead also by Ilya Sutzkever who transferred himself and his knowledge to Alphabet's LLMs competitor OpenAI quarreling in the process Alphabet's founder Larry Page and SpaceX founder Elon Musk who was then deeply involved with OpenAI. Some accuse Ilya Sutzkever of transfer of knowledge too copiously and abruptly from Google to OpenAI's Chat GPT's application, but nothing is proven. Nonetheless Google quickly caught by manufacturing under the umbrella of the parent company Alphabet the Large Language Models powerhouse Gemini, which works along with Bard quite well actually.
Amazon
Amazon is using AI intensively in its Amazon Web Services, AWS or cloud computing offering. AI algorithms are utilized within AWS to store, analyze data and produce actionable machine learning insights. Amazon is the leader in cloud computing, which actually is the application of AI and this provides Amazon with a significant competitive advantage. Cloud computing is a huge market by itself and Amazon AWS's dominance in cloud computing provides Amazon with edge in developing AI technology. In the last 5 years Amazon is developing very actively an internet advertisements business and now Amazon is the third largest company globally in terms of revenue from internet advertisements. Via internet ads Amazon gathers huge amount of information which together with the information Amazon gathers from corporate users of its AWS cloud offering could serves as a springboard for Amazon in AI software development. The AWS cloud tools recommendations are driven by AI, so the whole cloud computing business which arguably accounts for around 65 % of Amazon's market capitalization could be deemed an AI business. Cloud computing is very profitable for Amazon with a net profit margin of around 20 % compared with 6 % on average for Amazon's merchandise delivery business in the US. The international delivery business of Amazon is loss making. So the cloud AI business called AWS drives basically the majority of Amazon's business.
Meta
Meta Platforms, Facebook's owner is also in the AI Large Language Models, LLMs AI competition by developing and open-sourcing LLAMAs, its own Large Language Model based software algorithm. Meta, after Google, undoubtedly has the broadest span of user information, which if modeled correctly could be a treasure trove of insights to power user engagement on Facebook, Messenger, Instagram, WhatsApp and other Meta Platforms properties. Actually, open-sourcing LLAMA is a shrewd move by Meta, which will definitely be liked and attract following by the AI developer community. And make no mistake, whichever company has the largest number and highest quality software programmers writing code for its AI models, will be at the forefront of the AI modelling and ultimately computing race. Meta has many platforms on which to develop AI. LLAMA is actually quite an advanced Large Language Model, which could disrupt the LLMs space. Just as of 25th June 2024 Apple declined to further work with Facebook on Large Language Models citing data privacy concerns. This is a small upset for Meta, which will drive its own LLMs developement process.
NVIDIA
The poster child of the current AI boom is NVIDIA and its stock price. NVIDIA is currently valued at 2.25 trillion USD on the Nasdaq stock market. NVIDIA makes the graphical processing units, GPUs that power the AI servers. So basically NVIDIA is producing the picks and shovels for the current AI gold rush. NVIDIA has its roots in gaming, which high demand for computer processing power ensured NVIDIA has the know-how to build high producing, quality chips. In one interview some years ago, Jensen Huang, the founder and CEO said NVIDIA doubles the transistors in its chips or the quality of its GPUs every year. And that was some years ago. That is how advanced NVIDIA's production know-how is. Companies like ChatGPT, where Microsoft owns 49 % and Alphabet Inc, Google's owner produce software that makes AI, but they need server stations, which require chips, which NVIDIA and several other manufacturers like Intel, AMD produce. NVIDIA on the 6th of June surpassed Apple as the world's second most valuable company. NVIDIA's stock price has been rising parabolic for the last 2 years. As of 25th June 2025 NVIDIA is the world's most valuable company and the poster child of the current short-term AI bubble. In the last five years alone, NVIDIA's stock price has risen 20 times, which for a multi tens of billions of USD in market capitalization company 5 years ago is a staggering achievement. Currently, the demand for NVIDIA's AI chips seems insatiable. As of 10th July 2024, NVIDIA is the world's most valuable public company.
Tesla
Tesla is touted as the seventh member of the Magnificent 7 of AI technology companies. Tesla relies on the drivers' information it has acquired to produce self-driving cars in the not too distant future. Tesla has advanced AI know-how, but driving a car is extremely complicated and designing fully autonomous cars is a distant way off. Tesla has to achieve a breakthrough in electric vehicles production technology in the short-run, so as to produce electric cars profitably without tax subsidies. Meanwhile, its AI software engineers are working actively in developing fully autonomous driving technology. Tesla's intrinsic value is 120 billion USD, in Wolfteam Ltd.'s view. This reflects the value of Tesla's electric vehicles technology, artificial intelligence engineers, software engineers, financial professionals. The second quarter 2024 financial report by Tesla published the week ending on 2 August clearly showed Tesla's profit margins are decreasing. Profit margins after subsidies, that is. Without the tax and green gas subsidies Tesla is deeply unprofitable. Tesla's only chance to remain a going concern long-term is to achieve a technological breakthrough, which will make it profitable long-term, in Wolfteam Ltd.'s view. Thus Tesla could substantiate a market capitalization of 500 billion USD or more. The leading mass market automobile producers like Toyota and Volkswagen have net profit margins of between 2% - 4%, while luxury cars producers like Daimler, Audi, Porsche and BMW have net profit of between 4 % and 6 %. Ferrari, which is also publicly listed is an exception with its near 20 % net profit margin.
