AI music; the pros and cons explained, by AI!

Thumbs up or down for AI-generated music?

Whether you are aware of it or not, artificial intelligence (AI) is already impacting multiple areas of everyday life and work. According to Bill Gates – cofounder of Microsoft, one of the world’s largest and most successful software companies – the technology is nothing short of “revolutionary”, but as its usage grows it will likely prove to be highly disruptive, and have a major impact on an array of industries and jobs.

For composers, musicians, producers and DJs, it can be very disconcerting to discover that AI is already being trained and used to create, mix and master songs, produce “human” vocals, mimic existing artists, and even generate highly successful virtual bands. In an industry where many artists struggle to be discovered and earn a wage, the arrival of AI music could seem like a hindrance at best, and a career-ending catastrophe at worst.

As a composer and producer of tracks that are available to stream, download or license for commercial use, I wanted to discover how AI is already used in the industry, what it could mean for composers and producers in the future, how the law applies to AI manufactured tunes, and if there’s AI software already available that could benefit unsigned artists. In short, does AI music spell the end of true musical creativity as we know it, or could it purely represent a change in the ongoing evolution that is digital music production? And as this blog post is all about artificial intelligence, what better way to explore the subject than by asking an actual AI chatbot for the answers?

The following questions were all posed by Scambler Music on ChatGPT, an AI language model, and the quoted answers are generated entirely by artificial intelligence. The responses were only edited to remove duplicate or out outdated content, to correct language to UK spelling, and to include relevant links for reference.

This is a detailed in-depth post, and you can skip through the topics discussed by clicking on a question below:

• What exactly is AI?
• How is AI currently used in music production?
• How could AI impact the music industry in the future?
• As AI is increasingly used to create music how can composers remain relevant and well paid?
• Is there an argument for banning AI-manufactured music?
• Should music written by AI be classified as “artificial music”?
• What laws cover AI-generated music?
• For the benefit of human composers, can you suggest some AI-based production software that could enhance creativity?
• Are there any AI tools that musicians can use to better promote their work?
• Can you provide examples of successful artists that already use AI to generate chart music?
• In summary – a human perspective on AI chat, and what AI music really means for artists

What exactly is AI?

“AI stands for ‘artificial intelligence’. It refers to the ability of machines or computer systems to perform tasks that would normally require human intelligence to complete, such as learning, problem-solving, decision-making, perception, and language understanding. AI is achieved through the use of algorithms and models that allow machines to learn and improve their performance based on data inputs. AI is used in a wide range of applications, from self-driving cars and voice assistants to fraud detection and medical diagnosis.”

How is AI currently used in music production?

“AI is currently used in music generation in various ways, including:

• Music creation
  AI can generate music by analysing patterns and structures of existing compositions, learning from them, and then producing new compositions based on that knowledge. This can be used for composing new melodies, harmonies, and even entire songs.
• Music mixing
  AI can analyse and process different audio tracks to automatically adjust the levels, equalization, and other effects to create a more polished mix. This can help to speed up the mixing process and make it more efficient.
• Music recommendation
  AI can analyse users’ listening habits, preferences, and behaviour to recommend new music that they might enjoy. This can help users discover new artists and genres that they might not have otherwise found.
• Music transcription
  AI can transcribe audio recordings into written music notation, making it easier for musicians to learn and play the music.
• Music mastering
  AI can analyze and enhance the sound quality of a recording by automatically adjusting levels, equalization, compression, and other parameters to create a more balanced and professional sound. This can help to speed up the mastering process and make it more consistent.”

How could AI impact the music industry in the future?

“AI is already beginning to impact the music industry in various ways, and its impact is likely to grow in the future. Here are a few ways AI could impact the music industry:

• Music creation
  AI could help to generate new compositions and musical ideas, leading to an increase in the amount of music being produced. This could also lead to new genres and styles of music emerging.
• Copyright and royalty management
  AI could help to identify copyright infringement and ensure that artists receive fair compensation for their work.
• Live performance
  AI could be used in live performances to create new and interactive experiences for audiences, such as generating visuals and lighting effects in response to the music being played.

Overall, AI has the potential to transform the music industry in many ways, leading to new opportunities for artists, producers, and listeners alike. However, it is important to balance the benefits of AI with ethical considerations, such as ensuring that artists and creators are fairly compensated for their work, and that AI is not used to automate jobs that could otherwise be done by humans.”

As AI is increasingly used to create music how can composers remain relevant and well paid?

“As AI technology continues to evolve and become more widely used in music composition, human composers can take steps to remain relevant and well paid in the industry. Here are some suggestions:

• Focus on your unique strengths
  While AI can create music quickly and efficiently, it may struggle to replicate the unique style, emotion, and personal touch that a human composer can bring to a piece of music. Human composers can focus on developing their own unique strengths and abilities, such as their creative vision, musicality, and ability to communicate with clients and collaborators.
• Collaborate with AI
  Human composers can also collaborate with AI in order to enhance their creative output. For example, AI can be used to generate new musical ideas or to automate certain tasks, allowing composers to focus on more creative aspects of the composition process.
• Offer personalised services
  Human composers can offer personalized services to clients that are tailored to their specific needs and preferences. This can include custom compositions, live performance, or consultation services.
• Build relationships with clients
  Building strong relationships with clients can help human composers to maintain a steady stream of work and ensure that they are well compensated for their services. This includes being responsive, communicative, and collaborative with clients throughout the creative process.
• Continue learning and adapting
  As AI technology continues to evolve, human composers must continue learning and adapting to stay up-to-date with the latest trends and tools in the industry. This may include attending workshops and conferences, networking with other professionals, and experimenting with new technologies and techniques.

Overall, human composers can remain relevant and well paid in the industry by focusing on their unique strengths, collaborating with AI, offering personalized services, building relationships with clients, and continuing to learn and adapt to new technologies and trends.”

Is there an argument for banning AI-manufactured music?

“There is no strong argument for banning AI from music composition. In fact, AI has already been used in music generation with great success in recent years. AI can be used to create new sounds and music that would be difficult or impossible to create using traditional methods, and can even be used to generate entirely new musical styles and genres.

While some may argue that the use of AI in music may diminish the creative input of human artists, it can also be argued that AI can serve as a powerful tool for enhancing the creativity and expression of human artists. For example, AI can be used to generate new musical ideas that can inspire and inform the creative process of human artists.

Moreover, banning AI from generating music would limit the possibilities for innovation and creative exploration in music, and would stifle the potential for new and exciting developments in the field. Instead, it is more productive to embrace the potential of AI in music writing while ensuring that it is used responsibly and ethically.”

Should music written by AI be classified as “artificial music”?

“Music created by artificial intelligence could be referred to as ‘AI-generated music’ or ‘machine-generated music.’ However, whether or not it should be considered “artificial” is a matter of interpretation and perspective.

Some people might argue that since the music was created by a machine, it lacks the intention, emotion, and creativity of human-made music, and therefore should not be considered true ‘art.’ Others might argue that the AI’s algorithms and programming represent a form of creativity and that the resulting music should be considered a new form of art.

Ultimately, whether or not to refer to music created by AI as ‘artificial music’ is a matter of personal opinion and preference.”

What laws cover AI-generated music?

“The laws that cover AI-produced music can vary depending on the specific jurisdiction and the context in which the music is being used or distributed. Some possible areas of law that might apply to AI-constructed music include:

• Intellectual property law
  This area of law covers the ownership and protection of creative works, including music. If an AI generates music that is considered original and creative, it may be subject to copyright protection, and the ownership of the copyright may depend on the specific circumstances of the creation and ownership of the AI.
• Contract law
  If the AI-generated music is created as part of a contractual relationship, such as a work-for-hire agreement or a licensing agreement, the terms of the contract may dictate the ownership and use of the music.
• Data protection and privacy law
  If the AI-constructed music is created using personal data or sensitive information, there may be privacy laws that govern how the data can be collected, used, and shared.
• Consumer protection law
  If the AI-manufactured music is sold or marketed to consumers, there may be consumer protection laws that regulate how the music is advertised and sold.
• Competition law
  If the AI-based music is created by a company that holds a dominant position in the market, there may be competition laws that regulate how the music can be distributed and marketed to avoid anti-competitive behaviour.

In general, the laws that cover AI music will depend on the specific context and the jurisdiction in which the music is being used or distributed.”

For the benefit of human composers, can you suggest some AI-based music production software that could enhance creativity?

“There are many production software [tools] that currently make use of AI in different ways. Some examples include:

• Amper Music
  A platform that uses AI to create custom music tracks for various purposes like advertising, film, and video games.
• AIVA
  AIVA (Artificial Intelligence Virtual Artist) is an AI-powered composer that can generate music in various styles based on user input.
• Ozone 9
  A mastering plugin from iZotope that uses AI to analyze audio and suggest optimal settings for mastering.
• Magenta Studio
  An open-source platform from Google that uses AI to generate new musical ideas and provide tools for music composition and performance.
• LANDR
  An AI-powered mastering and distribution platform that uses algorithms to analyse and enhance audio tracks [this service was used by Scambler Music to master the tracks on the albums A world unknown and Radio Misophonia].
• Melodrive
  An machine learning music generation platform that creates adaptive music for video games, virtual reality, and other interactive experiences.

These are just a few examples of the many production software [tools] that currently make use of AI. As AI technology continues to develop, it is likely that we will see even more innovative uses of AI in song production in the future.”

Are there any AI tools that musicians can use to better promote their work?

“Yes, there are several AI tools that musicians can use to better promote their songs. Here are some examples:

• Soundcharts
  Soundcharts is an AI-powered music analytics platform that allows musicians to track their music’s performance across various platforms, including streaming services and social media. It provides data and insights that can help musicians identify their audience, track the success of their promotional efforts, and make informed decisions about marketing and distribution.
• Amuse
  Amuse is a music distribution platform that uses AI to identify emerging artists and predict their success. It provides artists with data-driven insights and promotional tools to help them grow their audience and increase their reach.
• Musiio
  Musiio is an AI-powered music discovery platform that allows musicians to promote their music to a wider audience. It uses AI to analyse music data and identify songs that match a user’s preferences, helping musicians to reach new listeners and increase their exposure.

These are just a few examples of the many AI tools that musicians can use to better promote their songs. By leveraging AI technology, musicians can gain valuable insights into their audience, create more engaging promotional materials, and reach a wider audience than ever before.”

Can you provide examples of successful artists that already use AI to generate chart music?

“There are several popular artists who have used AI to generate chart music or incorporate AI-produced elements into their music. Here are a few examples:

• Taryn Southern
  Taryn Southern is a singer-songwriter who has used AI to generate the lyrics and melody for some of her songs. In 2017, she released an album called “I AM AI,” which features songs created with the help of AI tools.
• Poppy
  Poppy’s 2020 album “I Disagree” features a song called “BLOODMONEY,” which was created using Amper Music.
• Holly Herndon
  Holly Herndon is an experimental musician who has incorporated AI-composed elements into her music. Her 2019 album “PROTO” features a choir of AI-manufactured voices that she created with the help of researchers at the AI research institute Jigsaw.

These are just a few examples, but there are likely to be more artists experimenting with AI in their music in the future.”

In summary – a human perspective on AI chat, and what AI music really means for artists

The ChatGPT answers on AI music were fascinating, well considered, intelligent, helpful and completely unbiased. It’s fair to say therefore that the AI language model operated successfully and provided useful insight on a complex subject. However, each answer was so balanced, with equal argument made both for and against every query, that the responses seemed to lack any real comprehension of human nature, and the likely outcomes of this new technological revolution (which probably won’t be as balanced as the AI answers). As a consequence, the feedback almost seemed a touch foolish, as if proposed by a very bright yet naïve child, with a blinkered or immature worldview.

Even when repeatedly pressed about the real threat that cheap mass produced AI songs could have on traditional musicians and composers, the AI chatbot never wavered – there are pros and cons and everything should largely work out okay. Whilst this type of answer is balanced and can seem reassuring, it lacks a fundamental appreciation, or acknowledgement, of the often flawed, unethical, illogical, unsustainable, and profit-driven practices of big business.

In addition to the utopian-like outlook offered by the chatbot, many of the positive sounding suggestions seem far fetched and unrealistic:

• In order to get discovered, would a talented yet overlooked guitarist, or jazz pianist, ever really pay for AI to provide inspiration?
• Are “building client relations” or offering “personalised services” the tasks that unsigned musicians should truly be focusing on?
• Would a folk group or metalcore band subscribe to AI-powered tech in order to alter their sound and increase their chances of success?
• Is there space in any traditional, classical or acoustic music genre for the addition of AI-generated effects or sounds?
• And is the aim of an experienced music producer to master a track to the best of their creative ability, or as fast and “efficiently” as possible?

One of the standout quotes from the chat with AI was:

“it is important to balance the benefits of AI with ethical considerations, such as ensuring that artists and creators are fairly compensated for their work, and that AI is not used to automate jobs that could otherwise be done by humans.”

This single statement summarises the biggest fear that many musicians have about AI music – artificial intelligence has the potential ability to generate more diverse music far quicker and much more cheaply than any artist possibly can. When the record companies that dominate the industry can use machine learning to generate new genres, songs, vocals, album art, music videos, and even entire virtual bands, how can unsigned artists possibly compete? Given the history of unfair and often exploitative practices that have plagued the economy since the onset of the industrial revolution, composers, producers and even music licensing businesses are entirely right to have legitimate concerns about their value, creativity, and place in a world with AI music.