AI will create 100x more new companies

I’ve been having this thought on how AI and how it’s going to create an almost infinite number of companies. The vast majority, the 95 to 99% of them will fail very soon. The reason is AI is allowing the ability to build any software extremely easily, 10 times easier than it used to be.
Here’s the rub and where humans come in. Just because you can build it doesn’t mean they will come. Humans are so important at understanding what it is that people actually need. Yes, AI will help with that a little bit, but I would argue marginally. It is humans and our ability to empathize with other humans, organizations, and companies that help us craft amazing products.
I think the next few years will be very interesting. There will be a lot of things learned. A failed company is really just a failed experiment. Ways to learn. But nonetheless, so many companies will fail. That doesn’t mean you shouldn’t try.
Right now I’m building a game with lovable.dev, a chrome extension with bolt.new, and business software with glideapps.com (yes they have AI too). Much of this is for me to explore and try out. But those who are extremely well versed in software development will be able to do much better, much faster. Those in product management and user experience design and user research will be highly sought after because they are the ones that will be able to talk to humans, organizations, and companies. Maybe this is not an unpopular opinion, but I thought it’s something I wanted to share.
What do you think? And I only used AI to transcribe my voice.