So we'll be going back to publishers as curators. Good for the publishers, I guess.
Earlier this year with a lot of luck, the Canadian duo Angine de Poitrine suddenly got discovered because they are doing stuff that falls outside of conventional music styles.
They aren't unique in the experimental nature they are exploring but it has highlighted an hunger from audiences to find stuff outside of the median. Folks like Frank Zappa had to relentlessly advocate for themselves as they figured there was a middle ground between these two thing.
even more consolidation and lock in
??
You say you deleted the tests, because you "should test it"? The logic seems inconsistent.
Sanity checking LLM-generated code with LLM-generated automated tests is low-cost and high-yield because LLMs are really good at writing tests.
I shipped a really embarrassing off-by-one error recently because some polygon representations repeat their last vertex as a sentinel (WKT, KML do this). When I checked the "tests", there was a generated test that asserted that a square has 5 vertices.
But LLMs let you skip all the boring parts - setting up harness, writing some initial inputs, adding asserts for every output. And then _you_ get to do actually important stuff, like ensuring square has 4 vertices.
I'm closely supervising the LLM, giving it fine-grained instructions — I generally understand the full interface design and most times the whole implementation (though sometimes I skim). When I have the LLM write unit tests for me, it writes essentially what I would have written a couple years ago, except that it tends to be more thorough and add a few more tests I wouldn't have had the patience to write. That saves me quite a bit of time, and the LLM-generated unit tests are probably somewhat better than what I would have written myself.
I won't say that I never see brain-dead mistakes of the "5-vertex square" variety (haha) — by their nature, LLMs tend towards consistency rather than understanding after all. But I've been using Claude Opus exclusively for while and it doesn't tend to make those mistakes nearly as often as I used to see with lower-powered LLMs.
No, they're absolutely shit at writing tests. Writing tests is mostly about risk and threat analysis, which LLMs can't do.
(This is why LLMs write "tests" that check if inputs are equal to outputs or flip `==` to `!=`, etc.)
Pisses me off on YouTube - it's really hard to find something genuine in the sea of the AI written, AI subbed, AI generated and AI published - it's a scourge not because it's there, but because the channels are lying about it AND because 99.99999% of what I encountered it's not worth the waste heat processing a "publish 100 catchy videos about current affairs".
Hard to believe these models won’t get better and better at producing music that humans want to listen to.
AI music I've heard universally sounds bland and robotic.