I did my PhD in SE after a couple of decades of working in industry. In my experience you can do a PhD in your 40s if you are motivated and have a worthwhile pr

I did my PhD in SE after a couple of decades of working in industry. In my experience you can do a PhD in your 40s if you are motivated and have a worthwhile problem to research and solve. But you probably won't get much support. Academia is disinterested in solving real problems experienced in industry. As you have discovered academia is driven by published papers, grants and the elusive concept of "prestige". Yes, AI is one of the more trendy areas for research at the moment. The reality is that having a PhD does make you over-qualified for many jobs. Earning more? Only if you live and work in a city where there are lots of startups and/or firms that seek out people with above average knowledge and expertise in specific areas. For example, some PhDs ended up working as quants in the finance industry. Bio-tech is another possible area. As for mental health, depression, imposters syndrome, etc are common. Less than half of PhD candidates complete their work and graduate. Sites like https://phdcomics.com https://phdcomics.com reveal how widespread the challenges are. As for financial challenges, that too comes with the territory. Don't take advice from people who don't have a PhD or don't hire those with PhDs. Reach out to alumni from your university and ask about their post academia experiences. In my case, I did the PhD for my personal interests. I had always wanted to do a PhD, but until I was established and able to self-fund, it was too risky. Clients hire me to solve their problems not because I have a PhD but because I have a track record in making money for my current and past clients.