From: cleoabram
AI is rapidly transforming the music landscape, offering new avenues for creativity and expression [01:09:00]. This new technology is viewed as a powerful tool that can help bridge “the gap” between an artist’s idea and their ability to realize it [03:43:00].
Shrinking the Creativity Gap
The “gap” refers to the difference between a perfect idea in one’s head and what can realistically be created [02:43:00]. Often, individuals may shrink their ideas to fit the available tools [02:52:00]. AI’s potential lies in reducing the technical skill required to express creativity, a historical trend that has led to more great music [03:08:00].
Historically, creating compositions required musical genius like Mozart or Beethoven, with significant costs associated with testing ideas with an orchestra [03:12:00]. Modern tools, like the computer for artists such as Avicii, have already lowered this barrier [03:38:00]. AI is seen as the next evolution of this, a very powerful new tool [03:43:00].
Current Applications in Music
AI is enabling the creation of entirely new songs through fake collaborations, or by generating lyrics and melodies in seconds [00:34:00]. Background music can even be created by typing in a few words [00:43:00].
Examples of AI-generated music include:
- Frank Sinatra’s voice singing Lil Jon’s “Get Low” [00:00:00].
- Linkin Park singing the “Pokemon Song” [00:11:00].
- Johnny Cash singing “Barbie Girl” [00:18:00].
Specific AI tools for artists include:
- ChatGPT: Used to generate lyrics [03:47:00].
- Google’s MusicLM: Used to create new melodies [03:50:00]. These may already be present on platforms like Spotify without listeners knowing [03:55:00].
- Elf Tech (by Grimes): Allows users to change their voice into Grimes’ voice for new songs and recordings [04:26:00], [07:19:00].
Artist Perspectives on AI Tools
Artists hold differing views on the appropriate use of AI.
Grimes
Grimes has publicly supported the use of AI, particularly voice cloning, becoming one of the first artists to do so [04:11:00], [04:13:00]. She teaches people how to use her Elf Tech tool for voice transformation [04:26:00]. On Spotify, there is a verified “human Grimes” and a “collaborator” profile for those using her voice [04:54:00]. She plans to release a competitive AI album alongside her own album [05:04:00].
However, Grimes believes that AI should not write lyrics or automatically generate music, as she thinks this might not be good for the human mind [04:03:00].
Spotify CEO Daniel Ek
Daniel Ek, CEO of Spotify, views AI as a natural progression of technological advances that democratize music creation [01:23:00]. He doesn’t believe Spotify would necessarily know what tools artists use to create their music, as they receive audio files [10:31:00]. Ek highlights that musicians have always learned from past artists [10:39:00]. He emphasizes that the output of AI training should be new intellectual property (IP), not something that “drafts off someone else’s name and likeness” or is “confusingly similar” [10:50:00].
Broader Implications for Expression
The potential for AI to aid human expression is significant [01:03:00]. While concerns exist about compensation for musicians [01:00:00], AI can be a tool to foster more connections through music [12:31:00]. The belief is that many people who may not realize they have musical talent could discover it through this technology [12:36:00]. This could unlock a vast amount of human expression currently “locked inside people’s minds” [12:56:00].