Can I Copyright AI-generated Images

Remember what we once read about in sci-fi novels? Technological marvels that can think, act, and create almost like humans?

Well, that's become a part of our everyday reality now. Artificial intelligence, or AI, is being hailed as a disruptive technological advancement.

Ever since ChatGPT was launched on November 30, 2022, AI has been all everyone seems to be talking about. But not all the talk has been positive. There are serious concerns with the rising popularity of AI. And these have sharply intensified over the last year. The most prominent of these has been about intellectual property rights.

Generative AI technology turns text prompts into creative output like text or images. The popularity of ChatGPT was the beginning of a wave of innovation in generative AI. But it was only one side of this innovation.

Generative AI for Images

Few realize that ChatGPT had an older sibling- a text-to-image AI art platform called DALL-E.

First released in January 2021, DALL-E predated not only ChatGPT but even other image-generation AI platforms like Midjourney by StabilityAI.

The interest in ChatGPT made users curious to explore more AI platforms and capabilities. But as the interest grew, so did the unrest.

A Statement of Policy from the US Copyright Office earlier this year said that technologies such as generative AI “raise questions about whether the material they produce is protected by copyright, whether works consisting of both human-authored and AI-generated material may be registered, and what information should be provided to the Office by applicants seeking to register them.”

As the US Copyright Office also acknowledged, "These are no longer hypothetical questions." That's true, but since these are somewhat complex questions, let's start with the basics.

What is a copyright?

A copyright is a form of intellectual property that protects creative works. This includes artistic, dramatic, literary, and musical works, among others.

According to the US Copyright Office, “Copyright is a form of the intellectual property law that protects original works of authorship.”

This sounds pretty straightforward. But AI technology has sparked a debate about what can be considered "original works of authorship." Now, for the first time, thanks to AI, the "author" can also be a machine.

Precedents for AI-generated image copyright

As AI technology is rapidly evolving, copyright law is evolving as well.

In 2022, comic book author and artist Kris Kashtanova won a copyright for her comic book "Zarya of the Dawn," But, in early 2023, The Copyright Office reversed this and reissued a partial copyright. Why? Well, they realized that the images in the book were created using the image-generating tool Midjourney.

"The Copyright Office concluded that a graphic novel comprised of human-authored text combined with images generated by the AI service Midjourney constituted a copyrightable work, but that the individual images themselves could qualify for copyright protection," the statement says.

In August 2023, a federal judge rejected a request to copyright a piece of AI-generated art. The plaintiff, Stephen Thaler, sued the US Copyright Office after his request for a copyright was turned down repeatedly. The work in question was an AI-generated artwork titled “A Recent Entrance to Paradise.”

Ruling on the case, Judge Beryl A. Howell of the US District Court for the District of Columbia stated that:

"Plaintiff can point to no case in which a court has recognized copyright in a work originating with a nonhuman," adding that “we are approaching new frontiers in copyright as artists put AI in their toolbox.”

Both these cases highlight one crucial aspect- the human authorship requirement.

What is the human authorship requirement for copyright?

According to the US Copyright Office's Statement, “Copyright can protect only material that is the product of human creativity. ”

It goes on to clarify that "the term' author,' which is in both the Constitution and the Copyright Act, excludes non-humans." Especially as more creators are experimenting with AI, the human authorship requirement will likely become central to copyright law in this regard.

While this provides a little clarity, the requirement is not absolute. Given that AI technology is constantly evolving, the US Copyright Office said, “This is a case-by-case inquiry.”

So, to what extent is creative-human involvement granted in the process for consideration? It raises the question, where does creative control end and AI begin?

What can I do to copyright an AI-generated image?

If you're applying for a copyright for work that has AI elements or involvement, there are a few parameters you can check to determine your chances of getting a copyright.

For instance, any output based on minimal human involvement won't qualify for a copyright. 

Avoid instruction prompts such as "arrange AI-generated material." Yet, essential AI-generated art might be further changed, enhanced, or edited by a human. So before applying, check the following:

1. Is your involvement visible and verifiable in the final output? 

If you've used AI to create certain parts or elements of your artwork, the final output should clearly show your creativity. You should be able to specify what parts of the final product were AI-generated. The Statement of Policy says that a human may "select or arrange AI-generated material in a sufficiently creative way tha t"the resulting work as a whole constitutes an original work of authorship." "

2. Have you sufficiently altered or enhanced the AI-generated content? 

In cases where your artwork is based on AI-generated content, the US Copyright Office will evaluate if an artist has modified the material originally generated by AI technology "to such a degree that the modifications meet the standard for copyright protection."

The disruption of generative AI

Generative AI is developing and disrupting technology at lightning speed. New applications and developments are coming to light almost every day and getting integrated into people's daily lives.

According to its website, Stability AI-backed research communities are “currently developing breakthrough AI models applied to imaging, language, code, audio, video, 3D content, design, biotech, and other scientific research.”

And it's not just Stability AI. Since ChatGPT's meteoric rise, tech giants like Google, Meta, and Microsoft have also been aggressively exploring AI capabilities.

This means that what we see today is only the start of an exciting new chapter in technology and innovation. There's much more to come, and copyright and trademark laws must evolve to keep pace.