Investing in AI: Everything you need to know

Artificial Intelligence (AI) has reached a decisive milestone with the emergence of ChatGPT. For the first time, the general public could experience the magic of a generalist AI capable of conversing, recognising images, and more. Meanwhile, AI has permeated the entire economy, from self-driving cars to drug discovery. As a result, financial markets are in a frenzy. Since late 2022, the stock prices of AI-focused companies, particularly Nvidia, have skyrocketed. How can investors navigate this landscape and invest wisely?  We provide the keys to understanding the stakes and investing in AI.

What is Artificial Intelligence (AI)?

AI refers to the ability of machines to mimic human cognitive functions to learn and solve problems autonomously. There are generally two main categories: narrow (or weak) AI and general (or strong) AI.

Narrow AI refers to systems trained to perform specific tasks. They often do so better and faster than humans. This is the most common type of AI today. Examples include search engines, self-driving cars, translation tools, etc. Their performance can be impressive but is limited to a specific domain.

General AI refers to a hypothetical AI that would be capable of performing any intellectual task that a human can do or even surpass them. This is somewhat the Holy Grail of AI, the equivalent of human intelligence that would have been reproduced artificially. But this remains conceptual today. No current system comes close, even if some recent feats of ChatGPT give the illusion of a certain versatility.

What are the trends and advances in AI?

When we talk about AI today, we are typically referring to narrow AI, specialised models. For example, ChatGPT is a conversational application: behind it, there is an AI model at work, specialised in natural language, namely GPT-4.

In parallel with GPT-4, which captivated the general public, several other impressive AI models have emerged, demonstrating the scope of possibilities:

• Dall-E (from OpenAI) for generating images

• PaLM 2 (from Alphabet) manipulating text, images, videos

• Claude 3 (from Anthropic) and Mistral (from Mistral AI) for natural dialogue

• Codex (from Microsoft) for writing computer code

• Galactica (from Meta) for analysing science
  

Why such a craze for AI in the media?

Although Artificial Intelligence (AI) has existed since the 1950s, it was indeed in late 2022 that it reached a decisive milestone with the general public. The arrival of ChatGPT was a real catalyst. Within a few months, ChatGPT had won over 100 million users, a record.

Other key factors reinforced the perception of witnessing a new gold rush of the 21st century:

1. Tech giants (Microsoft, Google, Amazon, Meta, Apple) have launched a frantic race to incorporate AI into their products. Microsoft drew first by integrating GPT-4 into its Bing search engine and Microsoft 365 suite.

2. Lively debates about AI reveal both immense enthusiasm and concerns about its ability to widen inequalities, spread fake news, misuse private data... Not to mention the risks of a superintelligence that would escape the control of its creators. A theoretical threat, but serious enough for experts, from Elon Musk to Geoffrey Hinton, to sound the alarm on TV.

3. Legislative projects, on which Europe is a pioneer with its "AI Act" draft regulation, which aims to set safeguards for AI models. But between precaution and countries' own ambitions, the balance is hard to strike.

Why invest in AI?

There is a very concrete reason why investors and financial markets have been raving about AI since late 2022: the prospects for growth and profits. PwC estimates that AI could generate up to $15.7 trillion in added value by 2030. That's more than China's current GDP!

AI market growth potential

Bloomberg Intelligence calculated that annual AI-related spending is expected to reach $1.3 trillion by 2032. But where will this money be spent? Where are the profits and opportunities being created? To understand this well, AI should be seen as a sector with three levels:

• The "capacity" layer: this is the technological foundation of AI, a bit like the foundations of a house. It includes companies that manufacture the essential components to make AI work: electronic chips, servers, networks... None of this would exist without companies like Nvidia and AMD (processors), ASML (chip machines), or Equinix (data centres).

• The "intelligence" layer: this is where the magic happens! In this layer, artificial brains are created that allow AI to think and learn, thanks to tech giants like Microsoft (partner of OpenAI), Meta, Alphabet (Google), or Baidu.

• The "application" layer: this is the most visible part for the general public. We are talking here about the products and services of our daily lives, created by companies that use AI, such as Siemens (healthcare), BYD and Tesla (automotive), or Meta (advertising and social networks).

Each layer has its own pace of development, and some companies (Microsoft and Alphabet) are present at multiple levels. This is a key point for the beginning investor.

Benefits of AI in various sectors

Within a few years, it will be difficult to find a profession that has not been transformed by AI in one way or another. AI will revolutionise all sectors, not just tech. Its impact will be profound and lasting on lifestyles:

• Healthcare: AI is revolutionising early diagnosis, personalised medicine, and drug discovery. Companies like Eli Lilly and Novo Nordisk are using AI to develop innovative treatments, particularly in the field of obesity.

• Finance: AI optimises fraud detection, scoring, and trading. Insurer UnitedHealth integrates AI into its processes to improve healthcare and the efficiency of its insurance agents.

• Industry: AI boosts productivity through predictive maintenance and automation. Leaders like ABB and Rockwell Automation are developing industrial AI and robotics solutions.

From an investor's perspective, these sectoral transformations create opportunities, both for AI players and for traditional companies adopting these technologies to remain competitive.

Direct investment: How to invest in AI companies 

To invest directly in AI, there are two key focus areas:

• Tech giants at the forefront of AI

• Chip and processor manufacturers

Tech giants 

Many consider the safest way to benefit from the rise of AI while limiting risks is to buy shares of large technology companies that already dominate this sector. Here are those that seem best positioned:

Microsoft (MSFT): The Redmond giant is the pioneer of AI for the general public thanks to its partnership with OpenAI. It integrates AI into all its flagship products, which has allowed it to record strong growth in its Azure cloud service. Microsoft is aiming for leadership in the "AI as a Service" market. This potentially offers a balanced risk/return ratio.

Alphabet (GOOGL): Google's parent company has colossal assets in AI, such as its mountains of data, its cutting-edge research subsidiary DeepMind and its top talent. Its PaLM 2 language model and Bard chatbot are highly performant. AI is expected to boost the relevance of its services and the effectiveness of its advertising. 

Meta Platforms (META): The social media leader is focusing heavily on AI to boost user engagement and open up new revenue streams. It is investing massively in generative AI, recommendation systems, and augmented reality. With over 3 billion users, Meta has a unique base.  

Amazon (AMZN): Contrary to appearances, Amazon is far from lagging in AI. The technology is present everywhere, from Alexa and recommendation algorithms to warehouse robotics. But it is especially AWS (Amazon Web Services), its global cloud division, that is democratising AI for businesses.

Apple (AAPL): Although very discreet on the subject, Apple has been integrating AI for years in tools like Siri or facial recognition. The Cupertino firm can count on two major assets: its ultra-powerful in-house chips and its closed ecosystem of one billion devices. Recently, the firm introduced Apple Intelligence. One to watch.

Other Asian giants like Baidu, Alibaba, and Samsung are also betting big on AI

Chipmakers

Let's review the main players in this segment and their investment potential:

Nvidia (NVDA): This undisputed leader in graphics processors (GPUs) for AI is benefiting fully from the "ChatGPT effect". Nvidia’s chips are essential for training data-hungry models and its revenue exploded by 400% in one year, causing its stock price to triple since January 2023. Despite an already high valuation, its growth prospects are dizzying thanks to its future generations of chips. 

AMD (AMD): Nvidia's main competitor, AMD is also heavily invested in AI with its Ryzen processors equipped with specialised cores. They are attracting more and more data centres. Less expensive than Nvidia, the stock still has a good margin for growth, even if the competition will be fierce.  

TSMC (TSM): This Taiwanese giant, the world's number-one semiconductor foundry, produces chips for Apple, Nvidia, AMD, and Qualcomm. The explosion in demand for AI chips ensures years of full-capacity production. It's like the Swiss army knife of AI!

ASML (ASML): With little known to the general public, this Dutch flagship equips the entire semiconductor industry with its machines, essential for engraving fine and high-performance chips. It has a quasi-monopoly on a critical link for the rise of AI. Despite a stock price that has soared, the demand is still there. 

Indirect investment: How to invest in AI-focused ETFs and funds

Investing individually in all AI-related companies can be complex for an individual. ETFs (Exchange Traded Funds) offer a solution by holding a portfolio of stocks. With a single ETF share, you own a fraction of each company in the portfolio.

Here are four ETFs for betting on artificial intelligence, each with a different approach:

ROBO Global Robotics and Automation (ROBO): This ETF invests in 80 to 100 companies involved in industrial robotics, drones, autonomous vehicles, and computer vision. It focuses on companies with "next generation" products in robotics and AI, rather than on mature technologies.

Xtrackers Artificial Intelligence and Big Data UCITS ETF (XAIX): Unlike ROBO, XAIX mainly bets on the big AI players, such as American (Microsoft and Alphabet) and Asian (Samsung) tech giants. This approach works well, with XAIX being among the best-performing AI funds.

Global X Robotics & Artificial Intelligence (BOTZ): Launched in the fall of 2016, BOTZ is one of the oldest funds specialising in AI. They invest in about forty companies active in semiconductors, medical robotics, and industrial automation systems.

WisdomTree Artificial Intelligence UCITS ETF (WTAI): WTAI seeks to replicate an original index, created to measure, which has the particularity of excluding American tech giants. The objective is to focus on companies that are purer and more specialised in AI, with potentially higher growth potential. It is an original and differentiating approach.

Venture capital and private equity funds targeting AI

For investors willing to take more risks and bet on future AI nuggets, venture capital and private equity offer another path. These funds invest upstream in unlisted startups with high potential and support them in their development, but with higher risk.

Access to these funds is often reserved for accredited investors, with substantial entry tickets ($100,000 or more). But some private equity giants like Eurazeo (RF.PA) in Europe or Blackstone (BX) in the United States are publicly traded. In parallel, Bitpanda also offers shares of behemoths like SoftBank (SFTBY), which launched a $100 billion mega AI fund.

Buying their shares allows indirect access to their expertise and their selection of AI nuggets.

AI projects in the crypto space

Finally, it would be a shame not to mention the growing links between AI and cryptocurrencies. Many projects are exploring how blockchain and tokens can be used to build a more open, decentralised, and transparent AI. We have selected four protocols that are building the bricks of decentralised AI:

Ocean Protocol (OCEAN): This "Nasdaq of data" allows researchers and companies to exchange their valuable datasets, essential for training AI models. This happens securely and is remunerated in $OCEAN tokens. With already more than 500 datasets, Ocean is a promising project anchored in reality.

Fetch.ai (FET): This project aims to create an economy where AI agents perform tasks and make decisions for users. For example, your Fetch assistant could search for the best price for a flight, book your vacations, and negotiate with other agents! The $FET token serves as fuel for this ambitious and futuristic ecosystem.

SingularityNET (AGIX): The objective is to create a network where different AIs cooperate to solve complex problems, like a global brain. $AGIX tokens remunerate AI developers. Partnerships with players like Hanson Robotics (creator of the Sophia humanoid robot) lend credibility to this project with an exciting vision.

Octavia (VIA): This AI virtual assistant helps manage crypto enthusiast and investor communities by moderating discussions, answering frequent questions, and providing trends. Its $VIA token allows access to its services—a simple and precise project.

Conclusion

To invest wisely in artificial intelligence, you can adopt a diversified approach in four areas:

• Focusing on AI blue chips: processor manufacturers like Nvidia and AMD of course, but also the obvious winners of this revolution with Microsoft, Amazon, Baidu and Google, who are democratising AI for individuals and professionals.

• Buying shares of companies integrating AI in a promising way in various sectors: Tesla (automotive), SAP (software), Dassault Systèmes (3D software), Airbus (aeronautics), and Samsung (connected objects).

• Aiming for broader exposure via ETFs specialising in AI and robotics, like GlobalX's BOTZ or WisdomTree's WTAI, offering a diversified basket of companies involved in AI.

• For thrill-seeking investors, taking an interest in cryptocurrencies linked to decentralised AI projects (Ocean, Fetch.ai, SingularityNET...), while keeping in mind their volatility.

The key is to maintain a long-term horizon and not put all your eggs in one basket. With curiosity and discernment, you are equipped to ride the AI wave.