
Search any artist’s exposure in AI training datasets, and a modeled estimate of what it’s worth AI ethics for creative work - music first. By @pescatios
Joined August 2026
- Tweets11
- Following39
- Followers6
- Likes111
onus retweeted
saw this last month and couldn't stop thinking about it. but i don't have a farm, so i made it into a game.
every video plays on a cow. 100+ cows to collect, free in your browser:
cinemooo.com
what's the first thing you're putting on a cow? 🐄
onus retweeted
i feel like even i’ve barely scratched the surface of what my own app can do.
upload any image, video, or gif. turn your favorite memes, memories, and music into your own beats.
listen to how they hummmmmmm
Onus makes part of an otherwise opaque system visible.
It’s hard to talk about consent, attribution or compensation if nobody can see where the work went.
onus.fm
If you’re a @lastfm user, come check it out / tell me what you think: reddit.com/r/lastfm/s/rUp44I…
onus retweeted
lol this can't be real. it's almost too on the nose.
earlier this week i somehow ended up at two very different events about the future of ai music on the same night. first was a Suno event at Loulou, right in the wake of their v6 launch. huge crowd, good food, DJs, a live coding performance from one of their ML scientists, stations to mess around with Suno. genuinely fun for a bit.
and to be fair, it was pretty much what they said it would be: meet the team, hang out, make some music, have a good time. but given everything that’s happened around Suno over the past couple years, especially recently, i was kind of expecting a little more engagement with the creative community around what this new chapter actually means. instead, mostly vibes lol. a little bored and honestly disappointed, i left and went to The Artist’s Stake.
xitter.cf/ednewtonrex/status/210…
This new lawsuit from Universal & Sony against Suno is big, because it makes an argument many have made previously outside of court: that training on ‘synthetic data’ should not get you off the hook for copying people’s work for AI training.
In the new lawsuit, the labels:
- allege Suno v6 was trained on outputs of previous “tainted” models (which themselves were trained on the labels’ music)
- say this amounts to “launder[ing]” [the infringement]
- expand the list of works they claim have been infringed to over 60,000 (from ~600 in their original lawsuit), which theoretically increases Suno’s maximum potential liability to around $9 billion
I suspect we will soon see more lawsuits around AI companies’ use of ‘synthetic data’.
Full complaint: musicbusinessworldwide.com/f…
Reverse-engineering the way it should’ve been
Earlier this summer, The Atlantic's AI Watchdog tool showed musicians which of their songs appear in public AI training datasets.
I built onus.fm to take that further: look up any artist, see how much of their catalog shows up, and get a modeled estimate of what that exposure could represent in dollars. Artists can search themselves, and fans can connect their listening.
758,889 artists. 13.5M recordings. Now with a modeled dollar estimate.
onus.fm
onus retweeted
Earlier this summer, The Atlantic's AI Watchdog tool showed musicians which of their songs appear in public AI training datasets.
I built onus.fm to take that further: look up any artist, see how much of their catalog shows up, and get a modeled estimate of what that exposure could represent in dollars. Artists can search themselves, and fans can connect their listening.
758,889 artists. 13.5M recordings. Now with a modeled dollar estimate.
onus.fm











