Sure, we've all known founders who've created products that were successful without talking to potential users, but it's very rare. What we've all encountered more often: founders who stubbornly don't talk to users as some sort of half-understood adherence to the biography of Steve Jobs. And then their product fails. If they had only tried a different way -- I don't know, talk to the people who might pay them? -- they could have avoided complete failure. It's not as glamorous as magically stumbling on a good idea, perhaps, and requires extra work, but it's often a more viable path for most of us.
I think what gets overlooked about Jobs, even among his admirers, is that he did start with an audience. Particularly by the macOS X and iOS days, Jobs was fundamentally driving the product design towards what he wanted with the implicit idea being that if he liked it and used it, others would too.
This has its own pitfalls, but he was fundamentally "talking to users", he had just narrowed his focus group to a user of one. He used his own products as far as we know as well.
The thing is that most tech doesn't dogfood quite so easily. Most founders can't replicate that as they are generally selling outside of the consumer market that they can just adopt themselves.
So, maybe cast another way, every product must have a target demographic and you need to interact with that demographic to get feedback on what you're building. Jobs just happened to have a shortcut that isn't generally applicable.
And of course, Jobs built the rest of Apple around this culture.
I read Ken Kocienda's book about working on the original iPhone keyboard [1]. He says development revolved around demos: first to each other, then to senior leadership, then to Jobs.
2000s Apple tried to hire people with good taste, then constantly tested against that taste. It was all internal, but it was testing.
