The harsh truth is that copyright for AI-generated content is currently a gray area - We simply don’t have clarity on whether it can be owned for marketing and branding purposes. There is another aspect around the ethics of AI-generated content, but that’s an entirely different conversation. The questions that arise here:
We don’t have answers to these questions yet, as this field is still in its nascent stage. Until these are answered, companies need to tread carefully.
The primary reason for these lawsuits is the use of datasets to train generative AI models – original art scraped from the internet without the consent or even knowledge of the artist/creator.
And there is some precedent for this, albeit in a different setting. Remember the Naruto vs Slater case?
David Slater, a British wildlife photographer, had left his camera unattended on the island of Sulawesi when a curious Indonesian black crested macaque, Naruto, clicked several selfies. Slater published these images in a book, but a dispute arose with organizations claiming that the photo was in the public domain due to the ownership lying with Naruto - as it was the monkey who clicked the photo and not the photographer.
However, the US Copyright Office stated that it “will refuse to register a claim if it determines that a human being did not create the work.”
This might be relevant to the Midjourney Class Action lawsuit where a group of artists is claiming copyright infringement. The outcome of this lawsuit could have far-reaching consequences for AI-generated content.
Let’s look at a few examples –
If the source material used to train the AI is in the public domain, then the resulting AI-generated content may also be considered public domain. This could mean that anyone can use or modify the content without permission from a copyright holder. Would the Office then consider it to be protected under the original copyright?
Since the 1980s, the UK has protected the copyright for computer-generated artwork as long as it has been created with AI assistance and human creativity. For AI-generated work, it clearly designates “the person by whom the arrangements necessary for the creation of the work are undertaken” as the author and gives full copyright to them.
But things have changed today, especially with AI tech. There is a chance that the UK government had anticipated the copyright issue around data used to train AI generators. Thus, they had proposed plans permitting organizations to allow data mining for AI training without being implicated in copyright infringement. The law is not yet in place though; we’ll have to wait and watch how the UK government plans to roll it out.
However, this has not passed yet and lawmakers are still tightening regulations around AI tech and generative AI.
If you are looking to use AI-generated art commercially, then you may want to consider these things:
Coca-Cola has partnered with Bain & Company and OpenAI in a bid to use generative AI for its marketing content and customer experience. They’re inviting digital artists to compete in the “Create Real Magic” contest, using AI platforms to generate original AI art from Coca-Cola's existing creative assets.
This is an interesting use-case as Coca-Cola owns the copyright to all their creative assets.
As AI technology continues to advance and become more widespread, it is likely that copyright laws and regulations will adapt to address the new challenges and opportunities presented. The future of AI copyright is definitely complex. However, the ongoing discussions and debate among stakeholders in the legal, technological, and creative fields is a good starting point.
Write to us and let us know whether you think there should be copyright provisions for AI-generated content or if it should be free use.