Has AI made it easier to build the wrong thing faster?

AI tools have made it much easier to build products, prototypes, landing pages, automations, and MVPs.

But I keep wondering whether this creates a new problem:

It is now easier than ever to build fast — but still very hard to know what is actually worth building.

For founders, indie hackers, freelancers, agencies, and product teams:

How do you validate an idea before investing serious time into it?

What signals do you actually trust?

For example:

• customer interviews
• pre-sales
• search demand
• Reddit or community discussions
• Upwork / freelance marketplace demand
• competitor reviews
• waitlists
• paid ads
• manual concierge tests

And what signals have misled you in the past?

I’m working on OPBO, a product in this space, and I’d love to understand how builders think about validation before launch day.