Ask HN: Why AI companies so limited?
5 piratesAndSons 4 8/18/2025, 10:13:37 PM
You’d think with all the hype and money flowing into these companies, they’d actually come up with something revolutionary.
Instead, we see yet another note-taking tool. Another rent-seeking middleman. Another iteration of something that already exists—just with fewer people employed, because costs were cut.
Where is the equivalent of building semiconductors for this era? Where is the leap from a horse to the Model T? And if that’s too hard, why not go after industries everyone hates—like U.S. health insurance? Where are the AI companies burning billions to make healthcare as cheap and accessible as an iPhone?
Setting aside the industrial copyright theft issues, it would be one thing if “AI bros” were actually delivering something that changed society for the better. But instead, it feels like AI is not only large-scale copyright theft, but also, for the most part, useless.
If you want to make an AI agent that helps people negotiate down bills? The insurance companies will not even deal with you unless regulatory-capture-voodoo requirements are met.
AI right now is mostly a 'low-hanging fruit' playground—copying and improving existing processes, automating small tasks—because those yield faster returns. That said, the potential for big leaps is still there; it just takes the patience, capital, and vision to go after the hard problems that actually move the needle.
As for why they are not delivering anything, it's IMHO another proof that AI cannot deliver without serious technical knowledge, which those spammers lack.
Most of those companies pretend that anyone can vibe-code, but it's obviously not true.
Now go find gold.
We are living in an era in which the expansion has dragged on long enough, and we are due a new expansion any time soon.
However, the game will not be the same afterwards. It might not even exist.