- cross-posted to:
- hackernews@derp.foo
- cross-posted to:
- hackernews@derp.foo
Google is embedding inaudible watermarks right into its AI generated music::Audio created using Google DeepMind’s AI Lyria model will be watermarked with SynthID to let people identify its AI-generated origins after the fact.
A needle in a haystack is much harder to find if the haystack is the size of a truck. People don’t have infinite time to listen to music, and if it’s almost all the same, they’ll stop trying to find upcoming artists, ai or not.
We need an ai to listen to music and tell us what to like by playing it on repeat till we do. Just like the radio stations.
I think probably the vast majority of up and comers in the music scene come from, not just randomly going viral (which I don’t think will be a concern with AI music anyways since it will probably sound like shit for the next 50+ years), but probably comes from just trolling around and doing local shows in venues that they know will attract that people who like the noise. I don’t think it’s very hard to distinguish between AI and people in that context.