Download Carl's Bear Market Survival Guide e-Book:
https://www.thinkmarkets.com/au/lp/2023-bear-market-survival-guide-ebook/
Next big thing
Fads come and go, and no doubt you clicked on this article because AI has been all over the media lately. I've seen many fads come and go over the past 30 years of my investing career. I made a million dollars in the Dot.com boom in 1999-2000 only to give it all back in the withering bear market which followed! So, yep, "
Been there, done that!"
Dare I say though, I don’t believe AI is a fad. It's not even new really. AI has been with us for many years as businesses have been looking for ways to eliminate (ahem, augment!) human inputs to the profit-making process. Humans, with all their wage requests, sick days, and infinite inefficiencies (read as obsession with watching cat videos during working hours!) are both expensive and annoying. They hinder the profitability process. According to business, if those pesky and expensive humans can be done away with – then zillions more in profits await!
Ok, jokes (and references to a certain 1980's movie starring a certain Austrian body builder aside for now), my advice is: AI's recent push into the headlines is something investors should take note of. The good news is, even after the recent spike in the prices of many AI focussed companies, I believe we remain closer to the beginning of this phenomenon/investment opportunity than to the end of it.
AI = 'All In' for tech industry
AI has been used for many years to cater to our basic needs. Think of Siri, Alexa, Google, etc. They automatically respond to our inputs based upon their programming. These are known as "Reactive" machines because that's really all they can do. They don't have any memory or ability to learn, let alone think for themselves.
The next step up is "Limited Memory" machines. This AI, also known as "Deep Learning" AI, has all the abilities of their reactive counterparts, but can also be trained to infer a more desirable outcome based upon historical data and trial-and-error outcomes. Sometimes, reactive AI gets it really right – as in when it uses previous and current satellite images to confirm power pole XF3456D1 has snapped in half and needs to be replaced, and sometimes it gets it really wrong – as in when you intend to write "gangster" in a text but autocorrect changes it to "hamster" instead…e.g., "That new Snoop Dog song is so
hamster!"
Limited Memory is where most current AI technology is at. It's trying to understand huge amounts of data to make better, more informed decisions about what the user requires. Self-driving cars are a great real-world application of Limited Memory AI. More recently, however, there's been a great deal of attention paid to a subset of Limited Memory AI called "Generative AI". This AI, as the name suggests, generates responses to user inputs. Chat GPT is an example of generative AI.
The difference between the AI required to drive your Tesla and Generative AI is substantial. Tesla AI, let's call it, requires substantial human training to teach it to recognise a stop sign from a crosswalk attendant. It responds to its environment based only upon what it has been taught.
With generative AI, the application will draw upon huge databases of information as a proxy for its knowledge and apply natural language processing (NLP) to converse with you about a topic of your choosing. Its next response depends on your next input, and so on. So, you could start with an empty input box and finish with enough python code to make your own version of Flappy Birds. You might have even seen some examples of generative AI in social media where content creators have used it to manipulate images, audio, or even video (breakdancing Biden anyone?).
It's generative AI which is probably commanding most of the attention at the moment, but the applications for AI in each of its forms is substantial, and there are massive amounts of capital being invested in developing each of the various AI technologies. I have chosen my Top 5 AI stocks based upon my view of which companies stand to benefit the most from the evolving AI revolution.
Top 5 AI Stocks in 2023
1. Nvidia (NVDA)
Nvidia supplies the high-performance computer chips which are essential in powering modern AI applications. In particular, it is the global leader in graphics processing units (GPUs). Deep learning models require a massive amount of computing power as they crunch through immense training datasets. Basically, AI needs to do a bunch of complex math, and GPUs relish math-based applications as they can process multiple streams of computations simultaneously.
GPUs are different from central processing units (CPUs), which are better suited to managing a computer's regular functional processes. Intel (INTC) is the world's biggest seller of CPUs and has only a modest GPU product offering compared to Nvidia. It's worth comparing Intel's price chart since the start of May when AI started hitting the headlines again, to Nvidia's. Clearly the market rates Nvidia's potential in AI over Intel's!
The other key growth area within the tech industry is also being driven by AI, is cloud computing and storage. Again, AI needs huge amounts of data, and huge amounts of data needs to be stored – preferably somewhere in the cloud for easy access. Nvidia is a leader in data centre hardware technology. As the cloud computing and storage sector grows to accommodate AI, Nvidia stands to be a major beneficiary.
Nvidia sparked this latest surge in the prices of AI stocks with the release of its first quarter 2023 earnings. Earnings per share for Q1 was $1.09, well ahead of analyst's expectations of $0.92. Even better, the company predicted second quarter revenues were likely to top $11 billion (vs analysts' expectations of $7.15 billion) specifically due to rapidly growing interest in Nvidia's AI chips.
Nvidia is clearly at the forefront of the AI revolution. Importantly for investors, it's not trying to develop any AI technology itself – it's simply supplying the industry with the tools to do so. It's a "
selling picks to the miners" model, and this significantly de-risks Nvidia as an AI play. They really don't care who wins the AI applications race, just that the race continues at fever-pace for a long time to come!
click to enlarge chart
2. ASML Holdings (ASML)
My second pick follows the bouncing ball one step back from Nvidia. Ok, so if Nvidia is selling picks to the miners…who's selling the pick-making machines to Nvidia? The answer is ASML. It's the world's leading supplier of lithography equipment used in the manufacture of semiconductors. That's fancy talk for "It makes the machines which makes the chips Nvidia and the rest of the semiconductor industry produce."
Investing in Nvidia certainly gives you exposure to Nvidia's growth outlook, but investing in ASML tacks on the likes of Taiwan Semiconductor, the world's biggest contract chip manufacturer, Samsung Electronics, and Intel, among others. Indeed, ASML is the dominant force in lithography tools accounting for around 90% of the global market in terms of revenue. Importantly, it's the only producer of the very specific extreme ultraviolet lithography tools which are used to manufacture GPUs – making it a clear AI beneficiary.
Not mentioned specifically here because I want to move past the "picks to the miners" theme in the next few selections, but certainly some key AI stocks to watch: ASML's major competitor Applied Materials (AMAT), Micron Technology (MU) which manufactures the high-performance memory modules essential for running AI applications, ASE Technology (ASX) which provides semiconductor manufacturing and assembly services along with semiconductor testing, and Advanced Micro Devices (AMD) and Marvell Technology (MRVL) which are arguably Nvidia's main competitors in GPUs.
click to enlarge chart
3. Microsoft (MSFT)
Microsoft is shaping as a major player in generative AI via its investment in OpenAI, the creator of ChatGPT. ChatGPT is a natural language processing-based software application which allows users to conduct human-like conversations about any topic of their choice. It is somewhat of a research platform upon which OpenAI aims to further develop its generative AI capabilities.
OpenAI was founded as a not-for-profit organization in 2015 by a small group of scientists, engineers, and some guy called Elon Musk. He contributed $50 million to OpenAI's early development and served as an initial board member. Musk left the company in 2018 due to what he explained was a conflict of interest due to Tesla's own AI research. He later criticized the company for transforming into its current "capped profit" model, and that Microsoft, which invested $1 billion in 2019 and $10 billion this year, has too much control over the company.
For Microsoft, the big carrot in AI is the integration of natural language processing technology into its suite of consumer and business software applications. The lowest hanging and likely most profitable route is to offer AI software as a service (SaaS) solutions to businesses through its Azure enterprise cloud platform. Next, and coming to a desktop near you, would be a range of AI productivity tools in Office 365 – imagine documents, spreadsheets, and emails which pretty much draft themselves. Finally, they'll be looking to add database mining and management tools via Microsoft SQL.
One thing is for sure, Microsoft has big plans for its upcoming AI integration rollout. Indeed, CEO Satya Nadella labelled AI the "
defining technology of our times". Even better, as a result of its relationship with OpenAI, Microsoft has a modest head start over major rivals Alphabet (GOOG) and Amazon (AMZN) who are developing their own natural language processing AI products.
click to enlarge chart
4. Palantir Technologies (PLTR)
Unlike Nvidia, ASML and Microsoft, AI is all Palantir does. Founded by Silicon Valley royalty Peter Theil, founder of PayPal, Palantir offers are range of enterprise software applications aimed at increasing productivity.
Customers include a range of blue-chip companies, but possibly the biggest stamp of approval for Palantir is that it counts the US Department of Defense (DOD) and the US Intelligence Community (USIC) as customers. The DOD uses Palantir technology to augment its operational decision-making capabilities, while the USIC (which includes the likes of the CIA and NSA) uses it for counter-terrorism analysis. The company was famously credited with a key role in helping locate Osama Bin Laden, and the CIA's venture capital arm In-Q-Tel is on the company's share register.
In the mainstream, Palantir products can help big business analyze huge swathes of data to spot anything from operational inefficiencies, cyber fraud, to the probability of a mortgage holder defaulting on their home loan. Importantly, Palantir is also well advanced in the generative AI race. Its new "AIP" product is purported to allow major organizations including the military to run large language models within their own private networks.
In an interview with Bloomberg Earlier this month, Palantir CEO Alex Karp indicated his company was experiencing unprecedented demand for its AI technology, suggesting the market for its products was "
infinite". He noted in-bound inquiries for the company's products and services had recently surged to as much as 12 times the usual amount. Karp also touted his company's AI prowess, particularly in the area of defense, boasting its products were so groundbreaking "
I'm not sure we should even sell this to some of our clients". In response to a question about whether Palantir's AI tech was potentially "
dangerous", Karp agreed, but warned "
Either we will wield them or our adversaries will".
(source: Lizette Chapman and Edward Ludlow, Bloomberg, Palantir CEO: AI So Powerful 'I’m Not Sure We Should Even Sell This'. June 2, 2023)
Given Palantir has the tech and connections to be the only company on this list to rival Cyberdine Systems aka '
Skynet', it's probably worth investing in it just to be on the right side of our future mechanical masters!
click to enlarge chart
5. Upstart Holdings (UPST)
I'm going to change tack a little with my final pick. So far, we've looked at AI software and platform providers, as well as the hardware manufacturers looking to cash in on the AI boom. Upstart holdings is a little different in that it's not doing any of these things, but rather it uses AI within the development of its own products and services.
The Upstart platform is a marketplace for personal and auto loans which connects borrowers to finance providers. It uses machine learning models to identify the credit risk of applicants more accurately than traditional methods which allows for faster and more tailored approvals. Upstart earns a commission on the loans which are written through its platform.
In addition to Upstart Marketplace, lenders and auto-dealers can white label the lending platform to offer loans directly to their customers. The application is cloud based and can be customised to each finance providers credit policies and internal IT systems to offer a streamlined end-to-end process of originating and servicing loans.
Upstart claims its AI technology is significantly more accurate than traditional lending models and leads to higher approval rates and lower default rates, thus benefiting both consumers and lenders respectively. Further, the system is constantly learning from the outcomes of tens of thousands of loan repayments vs defaults to constantly improve its effectiveness. Upstart stands to benefit as companies increasingly grow to trust, and therefore seek out AI-based solutions.
There are many other companies using AI to improve their products and services. It's worth checking out Adobe (ADBE) which is doing some amazing things with its Photoshop and Experience Cloud products, Salesforce (CRM) whose Einstein platform uses natural language processing to help business better interact with customers, and whose SlackGPT assists in managing and improving internal communications within a business, iRobot (IRBT) which uses AI in its suite of domestic cleaning products, and UiPath (PATH) which uses AI to power its automated robotics offering, and computer vision and cognition technologies.
click to enlarge chart
Conclusion
This Top 5 list, along with the 11 other companies mentioned within, should serve as an excellent launch pad for you to investigate opportunities in the AI space. May's huge pop in the shares of many AI companies (yes, including our Top 5) probably leaves many of you wondering if there's still any point in running out and buying shares in these companies right now.
Well, not even the smartest AI-powered application can predict the future just yet – so I will refrain from making a call with my non-AI powered brain on that one! What I can say with confidence, is we are very likely just at the beginning of the AI phenomenon, and in our understanding of its impact on how businesses run and on our personal lives.
Valuations may have been stretched a little in the very short-term on some of the companies mentioned above, but no doubt there will be pullbacks. As with any "next-big-thing", investors are best served by riding the momentum while it exists, but remaining on their toes, watching for signs the trend is changing. I regularly review the price charts and earnings outlooks of each of the companies discussed in this article for exactly this reason. You can join me weekly in my Live Market Analysis Webinars and follow along, plus ask questions about the stocks in your portfolio. Simply click on the link below to
register.
Learn More, Earn More!
Want your portfolio questions answered? Register for the next Live Market Analysis session and attend live! You can ask me about any stock, index, commodity, forex pair, or cryptocurrency you're interested in.