Six more August papers show where music models still fail
New work tests instruction following, song rewards, cultural understanding, audio fidelity, deepfake detection and the hidden layers used for music analysis.
What’s happening in AI music
Important releases, creative work, rights decisions and industry shifts — explained clearly and linked to the original sources.
Research
Four August papers test where AI-music evidence breaks
New studies examine musical sameness, broadcast detection, mixed AI stems and edited-audio false positives. Together they argue against one universal detector score.
Yandex Music reports replacing a 15-plus recommender cascade with Sona
One generative recommendation model beat the existing My Vibe stack in a smart-speaker test, according to a new technical report.
HappyShrimp targets full-song coherence with hierarchical planning and flow matching
The new song product arrives beside an Alibaba Token Foundry technical report built around global planning, local token generation and full-song audio rendering. The public paper does not itself name HappyShrimp.
Korin AI asks what music generation looks like when built in Nigeria
New research treats the platform as neither a simple success story nor a copy of Western music AI.
More to know
Five July papers move AI music beyond the one-shot song prompt
New work explored playable soundscapes, compensation signals, low-data training, structured composition and full-song rendering.