Investing in Artificial Intelligence (AI) Stocks is BS
We thought we’d have some time before the artificial intelligence bubble starts forming but it looks like things are getting a bit out of hand. We pointed out before the 5 biggest artificial intelligence startups (all unicorns) and remarked that now there are somewhere in the range of +1,500 startups that are selling some sort of “AI powered” solution. According to our favorite voice of reason, Anand at CB Insights, “it’s becoming clear that if you have a website, you are now AI“. Of course Anand’s company also puts out the most interesting “CB Insights AI 100” list which shows us that there are lots of “there” there. We couldn’t be more excited by the prospects of AI so now we want to “invest in artificial intelligence stocks”. The problem is, there really aren’t any.
Of course, you wouldn’t know that based on what all the pundits are saying. Every single time there is an emerging technology theme that starts gaining some traction, the likes of Motley Fool, Jim Cramer on CNBC and people of his ilk start rambling on about all the ways you can invest as retail investors. We like to call this “investing in everything with Google“. Don’t be fooled. Let’s take a look at what everyone is saying about “how to invest in artificial intelligence by buying artificial intelligence stocks”.
MONSTER Profits? Fcuk yes. Those are what you use to buy mansions and supercars with. We’re in. So here are all the stocks mentioned in the above articles:
|Forbes||Motley Fool||CNBC||CNNMoney||The Street||InvestorPlace|
We’re not surprised to see Google make the list. While the Company’s ad revenues dominate their bottom line, they are said to be ahead of everyone with Deepmind. They also offer an open source AI platform called Tensorflow. We’ve held Google shares since not long after their IPO. It’s a great company that’s not afraid to “fail fast” and “innovate or die”. Since they use AI now for search results, it has become key part of their business. Just remember that almost all their revenues come from selling ads presently.
Amazon may be using AI but let’s face it. You’re largely investing in an eCommerce platform with shares of AMZN. As for IBM, a small but growing portion of their revenues come from their “cognitive solutions” division which is mainly Watson (deep learning). In 2016, 22.5% of their revenues came from “cognitive solutions” so that’s fairly meaningful. We’ve held IBM for a while now but for dividend growth reasons. Another popular “artificial intelligence stock” is Facebook. We have a personal dislike of Facebook both as a product and a company, and hope that with no barriers to entry, someone else eats their lunch. They’re investing in AI but it seems like they’re more a beneficiary of AI. The same thing seems to hold true for Microsoft and Salesforce.
The last two stocks, Mobileye and Tesla, are plays on autonomous vehicles which are heavily dependent on AI. Mobileye was acquired by Intel recently and investing in Tesla is largely a play on electric vehicles at the moment.
The only real AI stock in the bunch is Nvidia with their GPUs that are a “picks-and-shovels” play on AI. The Company knows it too, and they’re largely positioning themselves as an AI company. Just look at their homepage:
If you bought Nvidia stock a year ago, you’d be one happy clam. They were the best performing stock in the S&P500 in 2016.
The key takeaway here is that while all these companies are working with AI, only Google (Tensorflow), IBM (Watson), and Nvidia (GPUs) seem to be making a real effort to capitalize on AI as a product offering. This leads us to believe that the only real AI stock that exists today is Nvidia – and you can bet all that future potential is being priced in at a P/E multiple of 44 (P/E is this largely useless metric we use in finance to make it sound like we know more about the market than we actually do).
Before we buy some Nvidia shares and call it a day, there are a few things we think need further exploration in future articles.
Firstly, every single company in the S&P500 can benefit from using artificial intelligence and automation. Industries like insurance are going to be winning early on since AI driven predictive analytics are just cleaning up. You can invest in big insurance companies like Chubb and be sure they are all over using artificial intelligence to run their business better. At least we would hope. Based on that premise, we’re going to do a future deep dive into the financials of S&P companies that are industry leaders to see what sort of productivity gains they might realize from headcount reductions and try to extrapolate this out into hypothetical future earnings. What we’re hinting at here is that maybe an S&P500 tracker ETF might just be the safest and best way to benefit from artificial intelligence. Intuitively, it feels that we’re on the verge of long overdue market crash, but what if we’re at the cusp of the biggest bull market the world has ever seen when AI replaces 80% of all service jobs?
Secondly, there may be lots of interesting short ideas here for more sophisticated retail investors. We’re still convinced that solar is what killed coal and that’s on our to-do list to investigate. We want to do some hypothetical scenarios about what sort of stocks might get decimated by automation and try to identify some interesting short candidates.
In the meantime, we’ll expect to see any day now the emergence of over-the-counter (OTC) stocks claiming to be involved in AI and we’ll be sure to out them for the scams they try to pull. We’ll also keep a close eye on the IPO schedule to watch for any possible artificial intelligence IPOs. Stay tuned and do drop us a note in the comments section below with your own two cents on where this whole AI thing is heading for retail investors.
Do you want to trade stocks for free? For the rest of your life? TD Ameritrade has completely eliminated commissions for stocks, ETFs (domestic and Canadian), and options. Open up an account with TD Ameritrade today and make brokerage fees a thing of the past.