Ok, let’s say #AI gets to a point where non technical people can build production level apps without needing any code or infra skills. I have serious doubts about this, but for argument sake.
What becomes the differentiator to make a business successful? If anyone can easily build an app that mirrors your company’s app, what is going to make customers choose your business?
product quality and relationship
same as it ever was.
That’s my feeling as well. I have a hope that this would turn tech businesses into small, local businesses where you actually know your customers and their pain points.
You're cooked if it gets to that point.
If it truly is that easy, companies will simply hire and develop in house versus outsourcing so they own the IP and have complete control. One of the major reasons why companies go SaaS is they get access to the specialized expertise their solution provider brings to the table.
What do you mean I’m cooked? And what hiring would companies be doing? For software engineers?
I agree on the point on SaaS for sure. But I think that expertise only becomes more important in this scenario.
Software companies would be cooked because AI would be hyper deflationary at that point. They would need absolutely massive scale to stay afloat and even then it would be a losing battle because AI will continue to evolve.
What expertise exactly? If everyone can develop production-level applications now thanks to AI, there isn't much value in that skill any longer.
Stepping through this and thinking of evolution, if we ever get even semi-functional AGI, it would truly be over because then it becomes a set it and forget it workforce. You'd set up a legion of AI agents, give them access to your company data and they'd likely be able to do everything soup to nuts, in real time responding to changes in the business.
Ya, I personally don’t buy that well will ever get to a point where AI can do everything 100%.
But to be clear, in my scenario I’m imaging AI good enough to build apps, but not necessarily knowing what to build and would need to be “prompted” by a user. In that case, the problem becomes who can envision the most compelling solutions to problems.
You could argue it would take AGI to get an AI like what I’m suggesting, and I think that’s fair, but I also don’t buy that a statistical model is capable of AGI (nor do I think we’ll get to a point of deploying apps with 0 technical knowledge).
Regardless, I do agree that tech companies are cooked. I have come to despise SV type companies and it would bring me joy to watch them fail.
