It’s hard to read the news these days and NOT hear about how artificial intelligence (AI) is impacting our lives. From images and content your read online, to law enforcement and health care, chances are AI has impacted your life already and you don’t even know it. Where is this all going? To find out, let’s hear from Ludwig APC, a long-time Barrel O’Monkeyz client who specializes in intellectual property law and privacy matters, on the timely topic of “ChatGPT and Other AI Tools…”
ChatGPT and Other AI Tools: What are the Implications for Intellectual Property?
First, a disclaimer: this article was NOT generated by ChatGPT or any other artificial intelligence (AI) tool. I did it the old-fashioned way. I spoke with colleagues, read timely articles, let my mind ruminate on the topic, and then employed the gray matter between my ears to form the words you are now reading. Of course, the fact that I felt compelled to make such a disclaimer is commentary itself on the state of affairs as AI tools are quickly graduating from being mere helpers to potential authors in their own right. Therein lies the issue. What’s next?
While I think we’re far removed from Terminator style cyborgs roaming the streets and subjugating humanity—at least I hope we are—it’s prudent for us to entertain thoughts about when it’s appropriate to use AI tools such as ChatGPT and when it’s not, and to consider the implications of using such tools, especially as it relates to intellectual property. No doubt, this topic will evolve in the coming months and years.
What is ChatGPT?
ChatGPT was created by OpenAI, an artificial intelligence and research company. At its most fundamental, ChatGPT is an AI-driven processing tool that allows people to engage with a chatbot to ask questions, write emails and essays, and even generate programming code. There have been several versions of ChatGPT released since November 2022. Currently, the latest version of ChatGPT is GPT-4, which has enhancements over its predecessors.
ChatGPT is in what is being called a “research and feedback-collection phase,” so it’s technically free to use. I say technically because, as we all know, anything that’s supposedly free to use on the internet, isn’t necessarily free at all. For example, OpenAI’s access to all ChatGPT interactions with users raises issues in my mind around the privacy rights of users, how that information is used, and whether it is being monetized. Thus, is it really “free?”
How is ChatGPT used?
One of the primary functions of ChatGPT is be a “conversational assistant.” In other words, it’s someone to talk to and ask questions. The assumption, of course, is that the information provided by ChatGPT is accurate and impartial. Its’ difficult to know for sure, which I suppose makes ChatGPT very human-like indeed.
ChatGPT and other AI tools are also used to generate writing, marketing materials, images, programming code, and even designs. In many cases, students have used it to generate term papers and essays, which has schools and educators taking the hardline by forbidding its use.
With so much content generation, naturally, many questions abound.
- Is the content really “new” or is it just derived from the previously created IP of others?
- Who owns the content ChatGPT and similar AI tools create?
What are the implications for IP?
In general, as long as any new content that’s created does not copy or use the work and ideas of others, then there can be no infringement. That seems straightforward enough, especially when you consider that sentient beings, like we humans, are known to have an original idea or two. Thus, we are capable of creating something new. But can that also be said about ChatGPT? Has ChatGPT reached the stage where it’s creating truly “new” content ginned from its own ideas and thoughts, or is it simply sourcing a wide variety of existing materials created by human authors and inventors and presenting it in a way that makes it difficult to trace original authorship?
That’s hard to tell – and certainly open for debate and legal precedent.
That said, when ChatGPT does create content, who actually owns it? Currently, copyright protection in the United States applies only to original works from human authors. It would seem then that, as of this writing, only flesh-and-blood people can claim copyright protection.
However, there are cases being litigated in the courts right now covering works created by autonomous AI tools. Thus far, courts have generally viewed AI-created work akin to being in the public domain or classified as derivative. Stay tuned, of course.
Things get really tricky when we pause to consider various scenarios:
- What if two different users use ChatGPT or another AI tool to create similar content that they then use publicly, in marketing materials, for example? Can one claim infringement or rightly send the other a “cease and desist” letter?
- What if AI-generated content proves libelous or defamatory? Who is liable for its use, the AI tool company, the AI tool itself, or the original user?
With so many unanswered questions, both from a practical and a legal standpoint as the courts have yet to set precedent, individuals and companies may want to think twice about utilizing ChatGPT and similar AI tools for sensitive business and creative purposes, not knowing ultimately where rules of ownership and accountability might end up.
For example, I read recently where a tech company had banned the use of AI tools by its engineers and programmers. The fear, of course, is that the company’s proprietary code and its designs—possibly even trade secrets—could be accessed by the AI tool company, exploited somehow by the AI tool itself, or even leaked to competitors.