Hey, it’s Brian. This article is for informational purposes and is NOT legal advice, cool? Onto the article…
Just the other day, there was some interesting legal news regarding artificial intelligence and the law.
OpenAI, the company behind the AI software Chat GPT, announced that this “chatbot” was able to pass the bar exam. And, not only that, it was in the top 10% based on its score.
No word yet if ChatGPT is now licensed to practice as a lawyer…
Stories like these are interesting to read about. But AI is actually serious business, with the potential for this revolutionary technology to change the world as we know it, in all aspects of our lives.
That includes your work as an online business owner, whether you’re an e-commerce entrepreneur, influencer, content creator, podcaster, vlogger, or whatever else.
AI technology is evolving so fast, that it’s hard to keep up with. And that’s especially true when you look at AI through the lens of the law.
More on that in just a moment…
First, what exactly are these AI programs?
In the case of ChatGPT, it is “trained” by having it review massive amounts of data, like webpages, documents, reference sources, and more… including legal journal articles and the like.
This allows the AI to respond to questions, and the technology means it can actually sound human when it does. Sort of like Google on steroids. Or a student cramming for a test. That’s why asking it questions on topics can generate quite impressive and comprehensive essays, news articles, and the like.
No wonder college professors are so worried about their students using ChatGPT for term papers! It’s actually banned in schools in New York, Los Angeles, and elsewhere, as well as in universities around the world, including Oxford.
In reality, AI is not ready to replace lawyers (or writers, journalists, etc.) because while it is good at responding to questions thanks to already having been given the answers… it is not capable of critical thinking, creativity, or reacting quickly to a new, unexpected situation.
And in some cases, it just completely makes up fake laws and cases.
Still, the rise of AI does bring up some questions that you must consider for your online business as this technology further develops, more companies (like Google, Meta, and Microsoft, to name a few) get into the game, and AI becomes more common.
Who Owns the Copyright for AI-Generated Content?
As I said, it’s unclear what the future holds for AI. But now is the time that you should start to learn about the potential risks to your online business, as best we understand at this point, as well as the potential opportunities.
One of the first things to consider is who owns a work created by AI?
I’m talking about written works, of course, but also images generated by AI, which you might have also seen online.
Like written AI content, images are created when a human user gives a prompt. This issue created some controversy last year when an artist used AI software to transform a text prompt into a work that actually won a contest for digital art. Other artists were quite upset!
Either way, so-called expressive works like these look like they could have been produced by humans – if you didn’t know any better. Can these works be copyrighted in that case?
The U.S. Copyright Office is already looking into this question and issued official guidance on the issue just recently, on March 16, 2023. The word for now is:
Only material created by humans can be copyrighted. So, nothing computer-generated, even if a human offers a prompt.
There is an exception for “partially” AI-generated works. This could be when an author or artist generates material through AI and then selects and arranges it. In that case, that specific arrangement, if done creatively, could be copyrighted.
It’s important to note that if you apply for a copyright, you must disclose anything that is AI-generated in the work. The Office will examine each application individually with regard to these issues before issuing a copyright.
You can find full details on these new rules here in this announcement from the U.S. Copyright Office in the Federal Register.
I do want to mention some of the other implications of this…
- If you use AI to create something for your business… and somebody uses something that’s similar (because they gave an AI the same “input”), you can’t sue them for ripping you off.
- If you plan to use AI-generated content or other works for your business, it’s your responsibility to make sure that the work does not infringe on a copyright. (This could very well be the case considering AI apps are trained on copyrighted works.)
This is not the last word on this issue from the Copyright Office as AI algorithms are trained on huge amounts of works authored by humans – and many of those works are protected by copyright themselves.
Some AI boosters claim that this practice constitutes “fair use” under copyright law. With fair use, brief excerpts of a previous, copyrighted work can be used in a new work for purposes of criticism, news reporting, teaching, or research.
It must done in a limited way and with a transformative purpose.
For example, it would be fair use for a music blogger to quote lyrics from a song in a post reviewing an artist’s new album. Parodies are another form of expression protected by fair use.
For more info on fair use, see this article on fair use for content creators.
However, many legal scholars disagree that having an AI absorb reams of creative works to spit out new works is protected under this doctrine. And many companies and creators are starting to feel the same way.
In fact, there are several lawsuits centered on this issue:
- There is a class-action suit filed by a group of visual artists against DeviantArt, Midjourney, and Stability AI claiming that these AIs are being trained on protected works – and that it constitutes infringement.
- Stock photo company Getty Images is also suing Stability AI for the same reason.
- The New York Times is also suing Microsoft and OpenAI, alleging copyright infringement, among other things.
These lawsuits are still ongoing. And there are more sure to come as AI becomes more common. It’s an evolving issue. One that you, as an online business owner, must keep a close eye on.
Using AI-Generated Works… in Your Work
On this issue, the law and rules and regulations are one thing. But you must also consider the ethical implications of using AI in your online business.
As an online business owner, if you start using AI-generated content, should you disclose it?
For example, say you use AI to write a blog post. Should you include that disclaimer? If you are hired by a client to create a unique artwork or design… is it okay to use something created by AI and not tell them?
Sports Illustrated received a lot of public pushback after it was revealed they were publishing stories using AI.
These are questions only you can answer.
But my recommendation at this point is to make those disclosures. And to avoid inadvertently infringing on a copyright, you could use AI as a starting point and then put your own touch on the final work.