Last year’s Raw Data Feel from Everything Everything was possibly the first release I’m aware of that made anything of an AI contribution. The band making use of it to help generate some lyrics, they’re tight lipped about what is computer generated and what is written by a blood and guts person. In all honesty it felt little more than a marketing gimmick at the time. AI in 2023 though is exploding as a hot topic, pushing up the agenda with the release into the wild of ChatGPT and Microsoft’s apparently abusive Bing AI chatbot.
Regular Mumubl.com contributor Meg Bolland drops into the Mumubl newsletter to give some views on the progress of AI in music and it’s possible future impacts.
Artificial intelligence is on the rise, and it has been for decades, but only now are we taking it seriously.
Until the early 2000s, AI was taboo, saved for dystopian novels and flashy sci-fi movies. Then the internet became mainstream. Now, we use it every day. From asking Alexa to play our favourite songs to Spotify curating our beloved ‘Discover Weekly’ playlists. Not to mention Google Maps, Snapchat filters, facial recognition. The list goes on.
AI simultaneously attracts a lot of media attention whilst needing to be talked about more, but properly.
No doubt you’ve seen the endless headlines saying AI’s first stop will be to take everyone’s job, and the second stop will be to take over the world… The usual conspiracy theory, clickbait culprits.
Currently, a large proportion of the AI conversation is about creativity. In particular, what AI means for the music industry.
So, what is the current position of AI in the industry?
It might surprise you how much influence AI already has in the music industry.
AI is integrated across the entire musical process, including creation and marketing to how we listen to music. As it can analyse an unfathomable amount of data, it can make more ‘informed’ decisions than humans, who wouldn’t have access to this type of information. For example, AI can examine an album, and pick a song as a single based on what would objectively be the most popular based on elements found in other hits.
Streaming sites, such as Spotify or Apple Music, use AI to understand and cater their services to listeners e.g. personalised playlists and smart recommendations.
Besides mainstream musical processes, many experimental projects are circling the internet. For example, ‘The Lost Tapes of the 27 Club’ is a fascinating application of AI. The Lost Tapes of the 27 Club was created to honour musicians that struggle with mental health. The ‘27 Club’ is a romanticisation of poor mental health in the industry, so this team used AI to create an album of music the artists might’ve created if they didn’t pass away so young.
Why are artists concerned by AI?
AI can create new music and remixes, it can write lyrics from a prompt and even push music to its limits and discover new genres.
Whilst AI can save artists a significant amount of time on tasks, many are wary of the application. One of the things that add to the fear is that this technology is still in its early days, yet it’s extremely advanced. We don’t know AI’s full potential.
The main concern is whether AI will replace human musicians, and take them out of a job. Whilst this isn’t an illogical fear and technology can do what humans do to a high standard, it will never completely replace the musician.
We can teach AI frameworks to mimic creativity, but it can’t necessarily be creative.
Creativity is largely inspired by emotion.
One of the main reasons why we love music (and all art) is to connect with others’ work, which is largely based on emotion. We turn to it in all situations. Whether that is when we’re in a good mood with our happy songs (*puts ‘Don’t Stop Me Now’ by Queen on*), or we harness motivation for the gym with hype playlists. The artists have felt these emotions, and represent that through their tunes. Hence why we love their songs, as we see their perspective and relate to them.
AI will not be able to experience these uniquely human things, and the music it produces will not have the same depth. We will always want and need the human touch in songs.
How can artists use AI to their advantage?
All creative people know the despair of having creator’s block when you just can’t get the ideas out, whether that is writing or producing. One way you can use AI to your advantage is to prompt it to come up with initial ideas that you can produce and put your ‘human touch’ on. Or, use it to help you mindmap or draft ideas.
One of the great things about AI is that you don’t need to be a ‘Jack of all trades’ to be able to create something amazing. For example, if your strength is writing lyrics, then AI can help you with production. So, you don’t have to spend hours upon hours learning, and can master your craft of writing.
Friend, foe or partner in crime?
When I heard of AI creating and writing music, I was skeptical.
Skeptical is the wrong word.
It felt dystopian.
The music industry relies on creativity and financial resources, and AI is a more cost-effective alternative to human labor. Since you can overwork AI without requiring payment, producers may opt to use it instead of humans.
Artists, the press and basically every other person on social media were saying the same thing. Further cementing my fears.
But, then I decided to do my research, and bit by bit my fear of AI turned into excitement.
I read articles about how AI will help artists with disabilities play instruments and create music. For example, EyeHarp, which is (the first) instrument that allows people to learn and play music with eye or head movement. Making the music industry more diverse, and ultimately more interesting.
Another way AI will make the music industry more diverse is by removing some of the class barriers. Music equipment and software are expensive, and unaffordable for many. After all, we all know the classic archetype of the ‘struggling artist’. So, this technology offers a less expensive option to make music, giving more people the opportunity to be artists.
Also, AI can save artists time on more boring tasks. Therefore, allowing humans to spend more time on the creative side, and focus on crafting even more unique tunes. So, a lot of artists are and will make AI their partner in crime. I think this is an exciting idea. Artists can skip hours of their time on tedious tasks, and put that energy into experimenting with lyrics and creating unique music.
The only thing we know for certain is that AI will continue to change the music industry, our listening habits and even the idea of what we consider an artist.
Nonetheless, I believe that AI will never replace humans in the industry. Music connects humans and has the power to make people fall in love, save lives and put a smile on their faces. These experiences and emotions cannot be understood or replicated by AI. It is the human experiences that inspire the most meaningful music.
Artists torture themselves for their art. That pain, desperation and ambition that we love to hear in music will never be found in an AI-produced song.
AI tools to try out
If you are interested in testing out this technology for yourself, check out these online tools: