Artifical Intelligence (AI) as an emergent technology has rapidly brought many new opportunities for the art industry. The possibilities to quickly create art seem endless. However, this also raises ethical discussions about the ownership and creativity of AI-generated art (Jobin & Ienca, 2019).
When researching different AI tools, I came across Splice; a cloud-based music platform with millions of free samples to help artists create new sounds. Recently, Splice introduced an AI-assistance tool that can automatically match different sound layers together with machine-learning to create a fluent and unique sample. The tool allows you to play around with different genres, beats-per-minute (BPM), and components. Out of curiosity, I tried this AI tool out. You can pick any genre on Splice ranging from disco fever to the “no filter” which randomizes samples from all music. The tool then automatically generates a starting sound, that can be customized with different components. After some trials and errors, I successfully generated a stack with the help of AI.
Testing this tool was an interesting experiment to discover the possibilities of AI, but the more important question it raises for producers is who or what should be credited for AI-generated music. This music is developed by data-based algorithm machines. If an artist uses pieces of AI-generated sounds, should it be the tool who is given credits or the artist? Perhaps some artists do not enjoy hearing this, but copyright law grants credits to the creator of the art, despite if this happens to be a machine. This means that fully or partly AI-generated music should grant copyright to the tool. Recently, an artist even failed to receive ownership for images generated by Midjourney based on the creation of their own comic book (Grant, 2023).
Furthermore, some worry that AI will overtake the art industry, and that art developed by humans will not receive the same value for faster and cheaper created AI-art. Others will argue that AI-music cannot even be considered art. This is a debatable topic and depends on individuals’ beliefs. Nonetheless, I do think human artists should be protected in some way during this AI-revolution. Splice for example aims to put creativity and artists first. In this case, the AI-tool is simply an assistant to help in the creative process, and not replace it. I suggest that this is the right direction organizations should take.
I am interested to hear opinions of others on this topic! Should AI be considered art and should machines receive credits for their creations?
Sources:
Grant, D. (2023, May 5). New US copyright rules protect only AI art with ‘human authorship.’ The Art Newspaper – International Art News and Events. https://www.theartnewspaper.com/2023/05/04/us-copyright-office-artificial-intelligence-art-regulation
Jobin, A., & Ienca, M. (2019). The global landscape of AI ethics guidelines. Nature Machine Intelligence, 1(9), 389–399. https://doi.org/10.1038/s42256-019-0088-2
Very interesting topic since I’ve actually been using Splice for the last couple of years now! Although I’m still not very familiar with the AI tool in there. The question you pose about who actually takes the credit in these situations is a tough one, and in my opinion, is based on how much the generated sounds influences the music itself. This is because if something is sampled from an older record, the sample must be cleared with the original composers unless it’s edited enough so that the sample isn’t recognizable anymore. So, my point would be that if AI-generated music is used for a little piece of a song or inspires certain elements then the credit should be for the producer himself, since in my opinion it’s no other than using samples from an old song. What would happen If it were fully generated by AI? I don’t really have an answer to that question…
I think this is an interesting subject. Especially since it indirectly talks about whether all music can be called art. I agree with the fact that an AI tool should be seen as a tool. An extra something that can help an artist make art (or a non-artist mak non-art). I don’t think AI is necessarily that much different than all the other tools artists use.
Thanks for this! Super interesting read 🙂 The first time I heard about AI generated music was last year. It was the producer of big names such as Lady Gaga who was there, center-stage, advocating for a tool called Soundful. I was I shock – how could someone so instrumental to the industry be willing to promote such an application? But indeed, his perspective was similar to yours, the music industry can draw inspiration from AI-developed beats, use, and sample AI-generated melodies and in general test new sounds much quicker than before. I tend to agree with this as well, after all, are generative AI applications in music really much more than smart synthesizers? I do, however, think it is – as with most applications of generative AI – high time that we revisit policies and regulations around ownership and useability of cultural goods.