General performance drivers, before the financials.
Use Factset
Summary of AI Sector
failure rate for enterprise.
Huge CAGR projected of
Lots of the market centralized to north america.
Asia is the fastest growing market
Drivers:
LLM
Datacenter & GPU
Cloud compute
M&A getting their hands on this
Major Players:
Nvidia
Cloud Compute
Microsoft
Amazon
Google
Chip Producers
Asics like graphcore
AMD
Intel
AI software
Individual stuff like (DermisAI), where you just slap AI onto another business model like that french guy did with this.
Thomas Gonnet
Niche solutions to specific problems like that one guy david bitton making an AI clip maker for podcasts and youtubers.
Things to look out for in the companies:
Regulations behind AI, and what you can train on.
Meta just trained off of 8tb of pirated books, so clearly right now there is no defined regulation on this. Just get the best model possible implying greater growth in the near future until a regulation hits and these kinds of moves are punished or it restricts the growth of everyone's products to essentially a similar level.
Monetization
Players may just be looking to get the most users first, and then worry about getting profit later which means more growth now and more users, so eventually when you start charging, you can charge a little to a lot of people rather than start off with charging and forgoing a bunch of early point growth.