Kanye West Producer Claims He Made 400 AI Vocal Generations
Two days before Vultures 2 was released in 2024, an anonymous producer says he was brought into the project on an emergency basis. His complaint in Los Angeles Superior Court alleges that he built at least 13 custom AI voice models, made more than 400 individual vocal generations and supplied source material using his own voice.
The producer claims his work appears on five Vultures 2 tracks. Beginning in October 2024, he says, he also programmed AI vocals for Kanye West’s solo album Bully. He alleges that neither project has been paid for and that streaming services do not list him among the production credits.
The suit seeks at least $110,000, citing breach of contract, unjust enrichment and violations of California and Los Angeles protections for freelance workers. These remain the plaintiff’s allegations. Pitchfork requested comment from West’s representatives and had not received a response when its report was published.
Electronic speech is not a new proposition. In 1939 Bell Labs demonstrated the Voder,*1 a machine on which a trained operator used keys and pedals to form intelligible speech. The apparatus sounded futuristic, but a human had to perform every syllable. The present lawsuit restores that awkward figure to view: an artificial voice still has a worker behind it.
The numbers change the scale of the argument. Listeners suspected in 2024 that West had replaced parts of his performance with a digital voice replica.*2 If substantiated, the complaint describes a production system: custom models, hundreds of passes, a human voice used as feedstock and its operator absent from the visible credits.
West presented Bully, released in March, as an album made without AI, despite having previously discussed using the technology in his writing and recording process. The question is no longer merely whether the lead voice belongs to West. The credits must also find room for the worker who built its digital counterpart. That line, as so often in the record business, is where the real court case begins.
- The Voder, or Voice Operation Demonstrator, was an early Bell Labs electronic speech synthesiser demonstrated publicly in 1939. It did not speak by itself: a trained operator shaped the sounds in real time.
- A digital voice replica is a technologically generated voice that can resemble a particular person closely enough to suggest that they recorded words or singing they did not perform.
Source: Walden Green, Pitchfork ↗