AI Images: Two Aspects of Authorship

I think of AI-generated images in two aspects: Are they works of art and are they copyrightable? (Is that a word?)

I wouldn’t consider an image I created through prompts to be truly my work of art. At some point I am relinquishing control and decision-making to an AI platform – even though generative AI would probably produce a more pleasing image!

I suppose I’m old school in this regard; I believe that an artist needs to control the decision-making and creation all the way through to the final product in order for the piece to be theirs. In the case of computer illustration, design, or motion graphics, the technology programs used are an aid, but aren’t making decisions for the artist. This is why I fight the fight to become better at drawing rather than spend much time creating images with AI.

Which one did I produce by hand? You only get one guess.

However, if someone saw an image I created with AI and wanted to use it in an advertising campaign, I probably would have a swift change of heart and consider it “my” image under my ownership and ask for renumeration for its use.Is that wrong? And is it wrong to create and use AI-generated images? No! But there are pitfalls in the commercial realm due to the sharing economy mindset where everything on the web is considered free to use, and most importantly, how the U.S. Copyright Office views AI-generated images. They have ruled in three different cases that AI-generated art cannot be copyrighted; protection is only available for works created by human beings. But what standard defines this?

In an excellent Ars Technica article, author Timothy B. Lee compares the issue to a similar situation when photography was invented. At that time, some people argued that photographs shouldn’t receive copyright protection because in their view a camera just mechanically captures images of whatever it’s pointed at.

In an 1884 case, the Supreme Court ruled that a photograph of Oscar Wilde reflected the photographer’s “original mental conception”. The reading of this case all the way to today is that even though a mechanical process captured the image, it reflected creative choices by the photographer and therefore deserved copyright protection.

In the case of AI, the Copyright Office has concluded that using AI to generate art is a merely mechanical process with “no place for novelty, invention, or originality”. But the similarity between these two cases seems obvious, at least to me – in both a human is making creative decisions that result in the creation of an image.

I would think these rulings are having a chilling effect on the use use of AI-generated images for commercial purposes. I think at some point the Copyright Office’s guidelines will have to change, as it is rapidly coming to the point that it isn’t practical for the Office to have to decide on this for hundreds of thousands of images. Not to mention that programs such as Photoshop are incorporating AI features.

What is your POV on use and copyright of AI images? Are you using them in commercial work even though they’re not copyrighted?

Note: No Generative AI was harmed (or used) in the writing of this article.