Pozalabs Leads Ethical AI Music Revolution with Proprietary Dataset and Artist Collaboration

December 9, 2024
Pozalabs Leads Ethical AI Music Revolution with Proprietary Dataset and Artist Collaboration
  • Pozalabs, a South Korean AI music startup founded in 2018, specializes in generating original music, lyrics, and vocals without relying on copyrighted materials.

  • This proprietary dataset includes over one million MIDI sound samples and 50,000 vocal samples, all created by a team of 15 professional composers to ensure high-quality music generation.

  • Pozalabs' tools, such as LAIVE and eapy, empower users to create music independently, reinforcing the startup's commitment to ethical practices.

  • In response to inquiries from the Society of Authors, Composers, and Publishers of Music (SACEM), Pozalabs detailed its ethical training methods.

  • Huh emphasized that Pozalabs' approach respects the creative contributions of artists and aligns with ethical practices in the music industry.

  • However, Pozalabs faces challenges competing against startups that utilize larger datasets of commercial songs without permission.

  • This proactive engagement may serve as a model for future partnerships between AI technology and the traditional music industry as discussions about AI's role in music continue.

  • CEO Wongil Huh confirmed that Pozalabs does not use any external music for AI training without permission, emphasizing their exclusive reliance on a proprietary dataset.

  • The company aims to demonstrate that the future of music can be a partnership between human creativity and AI enhancement.

  • The company firmly opposes the unauthorized use of copyrighted music in AI training, viewing it as unethical and harmful to the creative ecosystem.

  • To promote transparency and collaboration, Pozalabs has invited SACEM representatives to visit its Seoul office and expressed interest in developing artist-specific music generation models.

  • SACEM previously sent a letter to Pozalabs requesting they refrain from using SACEM-affiliated music for AI training without permission, following their advocacy for ethical AI practices.

Summary based on 3 sources


Get a daily email with more AI stories

More Stories