The promise of artificial intelligence is now becoming a reality. In this article, investment manager Tom Miedema highlights how leading companies are exploring, embracing, and harnessing the opportunities afforded by this evolving technology.
In 1950, British computer scientist and war-time codebreaker Alan Turing proposed a test for determining the existence of artificial intelligence. The general hypothesis was that if a human couldn’t tell whether they were communicating with an artificially intelligent machine or another person within five minutes of engagement, this was a sign of human-like intelligence in the former.
Turing speculated that in about fifty years from the time of the publication of his paper1, it would be entirely possible to programme computers of sufficient power to make them play the ‘imitation game’ so well that a human interrogator would have no more than a 70% chance of making the correct identification within the five minute period. Almost on schedule, the artificial intelligence (AI) journey has progressed to the extent that a growing array of leading companies across the world are exploring and indeed embracing the opportunities that this developing technology can bring to their businesses.
As a theme, AI has been dominating the equity market psyche. The main focus of attention to date has been on the build-out of AI infrastructure, with the likes of Nvidia, the maker of three key AI processors, seen as the provider of the ‘pickaxes’ required to dig for AI gold.
THE ‘COMPUTE’ ELEMENT
This element of the AI story has rightly thrown a variety of other semiconductor-related businesses into the spotlight. AI has supercharged the demand for semiconductors, with generative AI (Gen AI) requiring a substantially more computational power. This demand for leading-edge ‘compute’ – processing power – is particularly positive for Taiwan Semiconductor (TSMC) and Dutch semiconductor lithography company ASML.
TSMC is the world’s largest maker of semiconductors with a customer base which includes Apple and Nvidia. The company’s competitive advantage rests on a number of factors. It operates a pure foundry model, unlike peers Samsung and Intel, which eliminates potential competition with customers. The company has a record of strong execution, while high levels of research and development spending maintain its market leadership. Furthermore, as was made apparent by a recent Research team trip to the company’s headquarters and main manufacturing facilities in Hsinchu, the company benefits from Taiwan’s rich semiconductor ecosystem.
AI processor revenues currently represent 6% of TSMC’s overall revenues, but these, higher value chips are predicted to grow at a 50% compound annual growth rate over the coming five years and are currently forecast to account for over 20% of sales. In the recent second-quarter results, management noted the strength in AI-related demand which is currently outstripping supply. Additionally, AI-enabled devices (smartphones and PCs) are expected to materially increase the silicon content required for these devices, therefore boosting demand across other parts of TSMC’s businesses.
ASML’s EUV tools are the critical workhorses of leading-edge chip manufacturing
We also recently met with ASML in the Netherlands, and enjoyed the rare privilege of touring the clean rooms used in the manufacture of extreme ultra-violet (EUV) and High NA next-generation EUV tools. Each of these latter tools can cost up to US$380 million apiece, are the size of a double-decker bus and are the critical workhorses of leading-edge chip manufacturing, an endeavour which sits at the apex of human technological ingenuity. These machines help chipmakers reduce the number of process steps in high-volume manufacturing with significant reductions in defects, costs and cycle time, i.e. the length of time required to produce a semiconductor chip.
With its monopoly EUV position, ASML remains a linchpin of the Moore’s Law-esque continuation of the drive towards smaller, cheaper, more powerful and energy-efficient semiconductors. Moore’s Law is more in the way of an observation that the number of transistors on an integrated circuit doubles every two years. In that regard AI is driving advanced semiconductor demand, which is very beneficial for ASML.
DEVELOPING THE INFRASTRUCTURE
Beyond this ‘compute’ stage of AI development sit the infrastructure companies specialising in providing the foundational technology and services necessary for the development, deployment and scaling of artificial intelligence applications. These businesses offer platforms and tools that enable client companies and developers to build, train and run AI models more efficiently. It is worth noting that this is a technology still in its infancy, although Microsoft Corporation and Alphabet have developed leading positions in this area.
Microsoft provides the tools and services that customers use to build AI apps and services which will run on Microsoft’s Azure cloud computing platform. The company is a leading investor in OpenAI, and the collaboration between OpenAI and Microsoft Azure has been a game-changer, democratising access to advanced AI and accelerating the development and deployment of state-of-the-art models such as ChatGPT, DALL-E and Codex.
Microsoft Copilot and Copilot stack are orchestrating a new era of AI transformation
Furthermore, AI will be built into every Microsoft cloud solution through its Copilot application. Copilot is an AI-powered tool to improve productivity and creativity across tasks such as ideas generation and preparing presentations. As a result, Microsoft’s management is confident that this will be one of the fastest growing business areas in the company’s history. CEO Satya Nadella commented after the third-quarter results in April that “Microsoft Copilot and Copilot stack are orchestrating a new era of AI transformation, driving better business outcomes across every role and industry”.
Alphabet is establishing a leading position in the AI investment and application race, rolling out its own Search Generative Experience – which will be part of Google – that will craft responses to open-ended queries, responding to prompts with images and text, offering improved user experience and precision. It has multiple revenue opportunities including Gen AI cloud services, improved precision on Gen AI improved Search responses, and improved user experience and engagement on other services such as Maps.
Crucially, Google enjoys three significant advantages. In search channels, Google owns Android, which has over 70% of the mobile operating system market. In its product suite, Google has nine consumer-facing products with over one billion users, all of which are free. And then there is the brand. Google is a verb and is synonymous with search, and advertisers have years of experience placing high-traffic, high return-on-investment ads with Google via Search, YouTube, AdMob, Maps, Shopping and Travel.
HARNESSING AI
While the market is currently preoccupied with the build-out of AI infrastructure, a growing array of companies are harnessing AI to enhance their products, services and profitability. The AI ‘gold’ for these companies will be in generating incremental revenues, selling more products or services, expanding into new markets or developing new business strategies. Or they may be looking to improve the cost or capital efficiency of their operations – optimising spending and investment to generate higher revenue and growth while minimising costs. There is often a common thread to market-leading companies in that they tend to have more data and stronger digital capabilities, the trust of customers, more capital to invest, and have management teams which have been proactive in thinking about how to use AI.
Adobe has been investing in AI for more than 10 years and has extensive expertise and a sizeable set of curated content (Adobe Stock), data and technology. It recently announced Adobe Firefly, an AI-powered creative tool that will be integrated across its applications. Firefly uses Gen AI and simple text prompts to create higher-quality images with better composition, photorealistic details and improved mood and lighting. Adobe expects this will provide users with an easier to use, more powerful and personalised experience.
Trust is key
Germany’s SAP is an enterprise application software vendor whose customer base includes ninety-nine of the world’s top 100 companies and has over 299 million subscribers to its cloud user base. Many of the company’s enterprise customers are not ready to leverage the benefits of Gen AI implementation due to outdated technology infrastructure. In many cases, their first steps to AI utilisation will require an upgrade to their enterprise resource planning technology and database software and a shift to the cloud, and this will be beneficial to SAP. The ‘trust’ aspect is also key. For SAP’s enterprise customers, many of the most useful AI applications will require access to internal data and companies should rightly be very cautious as to which vendors should be able to access this data. This should provide a large incumbency advantage to SAP.
Inevitably, AI is finding its way into healthcare applications. Intuitive Surgical is harnessing AI to unlock the enormous automation opportunity within robotic surgery to improve patient outcomes. It has been at the forefront of robotic-assisted surgery for more than 25 years and, with the launch of its latest da Vinci 5 robot, looks set to maintain its leading-edge advantage. One of Intuitive’s greatest advantages over its peers rests in the enormous amount of data it has accumulated over this period. Intuitive teams are using AI and machine-learning technologies to provide insights that may help surgeons using robotics learn not only from cases of lower complexity, but also from more complex cases that address anomalies and complications.
Technological progress is rarely linear, and we expect the same to be true for AI
The American scientist and futurist Roy Amara famously said, “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” Technological progress is rarely linear, and we expect the same to be true for AI. There will be periods when the market will get over-excited in terms of usage expectations and the pace of adoption or monetisation, and for some businesses AI integration may be a slower burn than some of the more bullish proponents of the technology expect.
The initial enthusiasm over what will be a long AI journey has been driven by the demand for leading-edge compute and firms supplying AI infrastructure. Companies such as TSMC, ASML, Nvidia and Microsoft will play a critical role in the AI ecosystem for decades to come. But how fast will the technology be adopted? How will it change our lives? And what AI-enabled products and services are we willing to pay for?
The next transformational AI wave will be dominated by a growing list of companies that are identifying real-world use cases for the technology. Consequently, as long-term investors, a focus of our fundamental analysis over the coming years will be on how companies are employing AI to reach customers, more efficiently and more profitably, with better products and services.
Sources:
1 Alan Turing Comments – MIND: A quarterly review of Psychology and Philosophy, No. 236, October 1950
Important Information
This article is provided for general information only and should not be construed as investment advice or a recommendation. This information does not represent and must not be construed as an offer or a solicitation of an offer to buy or sell securities, commodities and/or any other financial instruments or products. This document may not be used for the purpose of an offer or solicitation in any jurisdiction or in any circumstances in which such an offer or solicitation is unlawful or not authorised.
Stock Examples
The information provided in this article relating to stock examples should not be considered a recommendation to buy or sell any particular security. Any examples discussed are given in the context of the theme being explored.