Morgan Stanley outlined three scenarios for the future of AI. In all of them, there was only one winner.
Cloud giants are present in each of the options, but there is no clear leader among them
Morgan Stanley believes that the outcome of the competition between open and closed models may be less important than the pace of AI adoption / Photo: Tada Images/Shutterstock.com
Morgan Stanley, one of Wall Street’s largest investment banks, has identified three possible paths for the artificial intelligence market over the next few years: the dominance of closed models, a hybrid market, and the rise of open-source neural networks. The list of beneficiaries varies, but in all cases, the companies that stand to gain are those that build the computing infrastructure for AI, according to Business Insider.
The Triumph of Closed-Source Models. If proprietary solutions maintain their edge in quality, security, and convenience, companies will continue to pay a small group of developers for the most advanced systems. Among other beneficiaries, Morgan Stanley cites Amazon and Google, as well as suppliers of chips, networking equipment, and electricity—including Nvidia and Bloom Energy.
A hybrid market. Morgan Stanley considers this scenario realistic. For the most complex tasks, businesses will use expensive proprietary models, while for day-to-day operations, they will rely on cheaper solutions that can be customized to their needs. Computing power will be distributed across cloud platforms, private data centers, and companies’ own hardware. This scenario benefits Amazon, Google, Microsoft, and Nvidia, as well as Datadog, Palantir, CrowdStrike, Okta, and ServiceNow.
The Triumph of Open Models. If such systems can match proprietary ones in terms of quality and usability, costs will decrease, and businesses will begin to adopt AI more widely—including in their own data centers and on users’ devices. Such an outcome would benefit suppliers of enterprise hardware, cybersecurity solutions, and edge computing. In this scenario, the winners would include Microsoft, the Chinese companies MiniMax, Z.ai, Alibaba, and Tencent, as well as Dell, HP, Apple, and, once again, Nvidia.
The only company to benefit under all three scenarios was Nvidia: demand for AI servers and networking equipment will remain strong regardless of the type of neural networks. According to FactSet, over the past three months, the overwhelming majority of analysts have recommended buying Nvidia stock. Sixty-one out of 64 analysts have assigned it “bullish” ratings—Buy and Overweight.
The cloud giants are also represented in each scenario, but the balance of power shifts. Microsoft stands out if open-source solutions prevail, whereas Amazon and Google will be in a stronger position if proprietary systems maintain their lead or the market becomes hybrid.
For investors, the outcome of the competition between neural network models may be less important than the pace of AI adoption itself. If that pace continues to accelerate, providers of computing infrastructure will thrive, Morgan Stanley concluded.
This article was AI-translated and verified by a human editor
