Around 5 years ago, I posted a similar thread about my lack of understanding on electric vehicles. It brought forward a lot of interesting perspectives on the bull and bear side and I learned a good bit.
[https://www.reddit.com/r/investing/comments/m60b98/i\_have\_to\_be\_missing\_something\_about\_the\_ev\_space/](https://www.reddit.com/r/investing/comments/m60b98/i_have_to_be_missing_something_about_the_ev_space/)
Today, I am posting a similar question about AI and how I dont understand the investments we are seeing in AI and more importantly, where does the ROI on this investment come from?
I dont think I am putting any opinion out there that isnt already known/debated, but my views on AI are pretty straightforward:
1. AI has the potential to revolutionize how we live and can disrupt countless industries and long term, will be a huge benefit for society
2. Like the Industrial Revolution, this disruption will not be met with cheers and excitement. People will be nervous about losing their jobs, wealth inequality being driven further and further, and a deep mistrust of institutions
3. A large majority of the AI investment by big tech is being driven by advertising revenue/cash flow. If we enter a recession or that advertising demand dries up, how much will these large companies be willing to leverage their balance sheets. Look at how much Zuckerberg already has with Meta.
4. Most of this investment will never see a return much less an adequate return on investment. Its very different than the dotcom bubble. With the innovation of some of these chips/hardware, some of this AI investment will quickly become obsolete.
5. From an accounting perspective, it feels like companies are understating expenses like depreciation/amortization which will further drive down profits. Plus, these business "relationships" you see between Open AI, Oracle, Nvidia, etc... is it just glorified ribbon cutting? Is any value being driven through these deals besides seeing the 5-10% bump in your stock price the next day? It seems the whole industry is mostly just signing deals between 5-6 companies and moving money around on a spreadsheet.
6. China is just as much in this arena as we are. Back in the dotcom era, American Exceptionalism drove a lot of the investment in the US in the late 90s because there werent great alternatives. In today's world, that isnt the case. You have a China competitor that has proven it cant just go toe-to-toe with you, but can straight up beat you. If this race comes down to building, the US will lose. We already have lost the great energy race of the 2010s and is there any doubt that China will do what China always does? Drive margins down, reduce profitability, and ensure only the fittest survive. In todays market, you have Chinese tech companies that can compete with the Google, Meta, etc.
7. History suggests that the dominant players today might not be the dominant players of tomorrow. Part of that is not innovating and getting beaten by a smaller and more nimble version. Part of it is making poor investment, rushing to a gold rush with no gold, and eliminating any flexibility you had.
8. I see dozens of companies throwing billions at this AI infrastructure - data center build, components to cool the data centers, chips, nuclear energy, etc. How many of these companies completely lose out and see a 90% draw down like the past cycles such as electric vehicles (Tesla aside).
9. We dont have the power grid in the US to handle this new generation of AI investment. I read and hear good things about expanding the grid, SMRs, and that's great, but does anyone actually expect us to build all of this energy? Im pretty sure China builds more energy capacity in a month than we have in years. That isnt an easy trend to reverse.
10. How does the US government help/hurt this? Increased state regulation? Limiting immigration of top AI researchers/entrepreneurs? Slow rolling electric power expansion? Picking winners and losers and disrupting the free market?
11. Will American companies focus on every day improvements to life or these content creating opportunities in AI. I saw Open AI launch a new thing with virtual content creating and that's great, but I have to admit, i asked myself, is that really the best the best minds in AI can do? Are American companies going to go for the easiest marketable and content creating areas of AI instead of the truly ground breaking areas? Will that area be dominated by China?
This has become kind of a rant, and I apologize, but I am just curious. How do you see the AI cycle and how are you playing it? Im looking at companies that have been beaten up due to AI fears and selling call spreads, and other option players to take advantage of these high flying companies. I never short companies like AI high-fliers.