Illustration created by Midjourney with the prompt “a computer in the style of Henri de Toulouse-Lautrec. –v 5 –ar 16:9 “
The human creative community, the U.S. Copyright Office, and AI-enabled creators all have different ideas about where copyrightable human creativity begins. Are you entitled to royalties if your work is used to help train a model? If a model is prompted to create something that is inspired by a bunch of different styles is that work “original?” If a human uses what they’ve learned to craft a prompt is the result of that prompt “original?” Humans learn by mimicking. In fact, human beings are the most successful mimics on the planet. “Watch Mommy!” or “Daddy will show you how to do it,” or “Repeat after me” are phrases we have heard all of our lives. You will learn about this subject by mimicking the words, acronyms and concepts written here – that’s why you are reading. Generative AI learns in a very similar way. Has it crossed a line? If it has, we’ve been walking up to the edge for decades. Let’s explore.
Art vs. Craft
Artists are among the best mimics. They are experts at copying the techniques that allow them to practice their art. We think of someone as a great artist when they can communicate an idea that is familiar to us but feels somehow new and exciting. Although there are many ways to describe art (and by art, we are discussing fine art, music, videography, film making, writing, etc.) let us use a simple description we can all agree upon: “Art cannot be ignored.”
This is opposed to “craft” which is pure technique and which we can easily ignore. In the distant past, the line between a creative artist and an expert technician was clearly visible to people schooled in the art. A good technician could fool most or all consumers of art into thinking that they had artistic talent, but to a professional, the differences would be quite clear.
In the more recent past, say since the 1960s, it has become much more difficult to make such declarative statements. One reason is that the proverbial “bar” has been raised to unimaginable levels. In order to become a modern, world-class professional in any discipline (athlete, musician, singer, dancer, fine artist, concert performer, entertainer) you literally have to spend all of your waking hours pursuing that goal from early childhood.
Learning to Play
As it does in many aspects of the human endeavor, the law of unintended consequences here too plays a mischievous role. Due to the immense amount of training required to just be average, it is extremely hard for laypersons to appreciate the level of technique and technical achievement that most professional artists have today. But, do not be fooled by the fact that you do not understand — or can’t relate to — the content that new, young artists are producing. A young adult entering the arts today, with a hope of being successful, will have blinding, extraordinary technique. So much so that the average person would mistake it for artistic talent.
This is such an important concept, it needs to be repeated. Today, most people who are not schooled in the art cannot tell the difference between someone with true artistic talent and someone with extraordinary technique. In fact, the time, dedication and perseverance required for world-class technical achievement is so respected in our culture, that some truly talented professionals laud lesser talents because of their technique.
Before 1986, all musicians who learned to play musical instruments did so by trying to copy (as exactly as possible) the way their music teachers played. That’s right — they listened to a note and did everything in their power to copy it exactly. Intonation, quality, volume, attack, decay, sustain and release — every aspect of every note played was to be copied as exactly as possible. If you could put some notes together in a sequence, you could play a song. If not, you went out for a sports team.
Although we all have many, many similar attributes, we – each of us – are uniquely gifted with some level of intellectual and physical skills. No, we are not stating the obvious. Unknown to most, acoustic musical instruments do not automatically play in tune. This is not the time or the place to explain a well-tempered Western musical scale, but suffice it to say that the physics of acoustic musical instruments actually prevent the design and manufacture of any wind or stringed instrument that can inherently play in tune.
Every acoustic musical instrument has several inherent physical barriers that must be overcome to master it. These include use of your embouchure (how you use your mouth on a woodwind or brass instrument), motor memory (how you use your fingers to manipulate the instrument) and a complex set of mental skills (including hearing) which are not well understood. How a musician overcomes these obstacles is unique to the individual and it has a name: musical style.
There is another aspect to the creation of music … notes to play. Where do they come from? In our Western culture, the basic building blocks of music are the 12 notes of the well-tempered chromatic scale. After you learn to play all of them across the entire range of your instrument, you can play anything that has ever been written.
What does this have to do with musical composition? Everything! Putting notes together in combinations is also a learned behavior. We learn to play a song exactly the same way we learn to play a single note – by listening and copying. When you pen your first musical composition, you are paying homage to the work of every composer and every musician you have ever heard. The nature of musical composition and performance is that it is the highest form of mimicry. No one could ever become a world-class musician or composer without playing or writing things they had heard before.
One could easily argue that no living musician has ever played an original combination of notes, regardless of their level of technique or their isolation from other musicians. Why? Because in order to learn to play, you must copy someone else’s work product.
So now, let’s talk about a sixth grader in 1986 who has been taking music lessons in school since the fourth grade. She can play many familiar songs and communicate with her musical instrument and her relatively newly acquired skill of mimicking other musicians. If she is being trained as a classical musician, she is judged by how exactly she can play what has been written more than two centuries ago. If she is learning to play popular music, she is judged by how much she sounds like the original artists she is copying. No matter how we judge her musical ability, it is in the context of her personal ability to copy (as closely as possible) work that already exists.
Now, let’s say that this young lady is truly extraordinary. Let’s imagine that she is musically mature beyond her years and can already combine some of the notes and musical phrases she has heard and mastered into a new groups of notes and musical phrases. If this is accomplished seemingly at will, she is said to be able to improvise. If this is done by writing the notes and musical phrases down for others to play, she is said to be a composer. Regardless of her skill level, she can share her interpretation of the work she has learned to copy by playing it for you live or teaching it to other musicians so they can play it for you.
Sixth graders who demonstrate this type of musical acumen are lauded and paraded around as child prodigies. Under our copyright laws, she is prohibited from selling, distributing or reproducing an exact copy of someone else’s work and calling it her own. But, if she creates a work that is “inspired by” as opposed to “ripped off from” music she’s heard, no one is going to bother her. To be sure, there is a certain amount of craft required to communicate the “feel” of a song without copying it note for note. But to a trained composer/producer/musician (even one in training), it’s all in a day’s work.
Decades later, things have radically changed. Today, a sixth grader does not have to spend any time learning to master the physical limitations of acoustic musical instruments – music can be made with ubiquitously available software. These apps assist in music-making. People can control them, but the programmers and engineers who created the computer programs endowed the software with the ability to reproduce absolutely any sound that can be heard. These programs are not synthesizers that create synthetic sounds of acoustic musical instruments, these programs are digital audio recorders that make exact duplicates of other sound recordings and allow the user to manipulate them. The software presents no physical limitations for users to overcome, although there are still those pesky intellectual limitations (including hearing) that have a dramatic impact on the outcome of the work product.
The sixth grader who opens and starts to use a digital audio workstation or digital musical instrument makes music exactly the same way her ancestor made music decades earlier – with one notable exception. Instead of trying to copy the notes that she hears or trying to copy the musical phrases, she simply imports the existing recordings and manipulates the recording into a new work by adding her own unique signal processing, other sounds, and maybe even a few well placed original notes. It’s called a remix or a mash-up, but it should not need a new name; it is a unique, creative musical work – the musician is simply using computers to create with.
In the present day, these tools not only incorporate AI, they may rely on it for their core functions. And, of course, with generative AI, all our sixth grader needs to have is a musical imagination and a way to describe how she feels in words. AI will do the rest.
There is, however, a profound difference in the way established artists and our existing copyright laws view sampled mashups. Using actual sounds from other peoples’ work is not considered original – and there is no practical way for it to ever be so considered. As long as our sixth grader plays this work for her family or close circle of friends, she’s fine. But, far from being considered a prodigy, she is instantly considered a plagiarist and a thief and there is no commercial future for her work. (On TikTok or YouTube there might not even be royalties for the original works the derivative work was made from). And, the Copyright Office has stated that AI generated art is not copyrightable.
Back to sampled mashups, if our modern-day sixth grader decided to take a bunch of video from her collection, some video output from her Xbox or Playstation and mash it all together with some music and add a comedy voice over, her friends would call her a genius. She could post her work on commercial sites (like YouTube and TikTok) – then she would be a criminal. This type of video mash-up violates the rights of so many stakeholders, it’s hard to cite them all. The music rights, unions and guilds, and rightsholders of all kinds would all have a claim to this as a derivative work. It would be practically unclearable under our current laws.
Not only do we have a problem with the recorded music and video industries’ lack of vision with regard to having a legal way to upload and download their work products, we have a much bigger problem in that the same laws practically prohibit new, creative talent from exploring and growing their art.
Enter AI. Now, she can just tell it what she wants and her AI-co-creator will produce it for her.
From copying my music teacher to copying the great composers to mashing-up great recorded content to learning to ask AI to create – it’s all the same process. We are copying what already exists and trying our best to do it so others will consider it art – not craft. Today, the lines between original and blatantly copied or computer assisted or computer generated works are so blurry, the concept of an original work may need some adjusting.
The idea that almost any individual who feels something can use technology to communicate their feelings to others is magical. We have given this generation new tools to create with. This democratization of the arts and creative communication deserves a rip-roaring round of applause.
Author’s note: This is not a sponsored post. I am the author of this article and it expresses my own opinions. I am not, nor is my company, receiving compensation for it. This essay was partially excerpted from Television Disrupted: The Transition from Network to Networked TV (Focal Press 2006; 2nd Edition, York House Press, 2008).