AI Can Tell Us Something About What It Means to Be Human
How artificial intelligence is learning from the humanities — and vice versa
When we talk about artificial intelligence, we often focus its ability to learn, make connections and solve problems all on its own. But for all its independence, AI technology remains entwined with the people who create it, use it and live with its impact.
As artificial intelligence becomes a fixture in our lives, two doctoral candidates and a recent alumna from the University of Florida College of Liberal Arts and Sciences had a chance to examine how our society shapes the technology — and how the technology is influencing our society in turn.
At a recent conference hosted by the National Humanities Center called “In Our Image: Artificial Intelligence and the Humanities,” the three rising scholars explored the possibilities AI presents for their fields. Just as urgent, though, is the role the humanities will play in developing AI for a more just world.
“The conference helped me see that collaboration can be done across the humanities and the sciences, and how important it is that we create opportunities to do this,” said LAUREN BURRELL COX, a PhD candidate in the Department of English and program coordinator for the Center for the Humanities and the Public Sphere. “We both need each other.”
Their attendance at the conference was made by possible by residencies at the National Humanities Center funded by professor BARBARA MENNEL’s Waldo W. Neikirk Professorship, awarded by the college. Below, read about their reflections on the opportunities and challenges artificial intelligence presents for the humanities and the world we live in.
Solidifying a space for the humanities in the future
What was one thing you took away from what you heard at the conference?
This conference made me realize that there are a small yet growing number of scholars examining the impacts and unintended consequences that accompany AI’s rapid development.
Not only will this scholarship solidify a space for the humanities in the future, but it also acts as a way to thoughtfully and equitably develop systems that can aid all communities rather than the elite few.
What challenges does AI present for the humanities, and what opportunities does it present?
AI has already presented itself as beneficial to our society as AI systems have mathematically and algorithmically been programed to assist in a wide range of tasks. Unfortunately, AI has also proven to be biased, reinforcing the status quo, and in some cases being outright racist and sexist.
We impose our own values onto the systems we create whether we mean to or not. Identifying our own morals, values and beliefs as well as our own biases and downfalls, may help us to rewrite the future as we explore and create new systems and technology forms.
AI rooted in storytelling and dialogue
What interests you about the impact of artificial intelligence on the humanities?
While many people think of AI as being purely technological, as a computer science tool, AI is often rooted in storytelling and dialogue. When coders design chatbots, for instance, they often collaborate with writers because those chatbots need to have authentic-sounding voices when they interact with users. Without this humanities-based influence in designing AIs, chatbots might not be such a powerful tool in the mental health industry or in the customer service industry.
What problems can arise from the use of AI?
Unfortunately, AI is coded by humans and it perpetuates the biases that humans often have. For example, Google Translate has had a nasty habit of automatically gendering certain terms when a user is translating a phrase from a language like English into a more gendered language, like Spanish or French. The app may make faulty assumptions about the user’s gender by translating stereotypically “feminine” activities (like cooking or cleaning) into a feminine form or translating “masculine” activities (like owning a business or being a boss) into a masculine form.
Even more dangerous, perhaps, is how AI can misidentify and target BIPoC (Black, Indigineous People of Color) people in criminal cases. For example, when courts or police use AI to try to identify potential suspects or predict the types of people who are likely to commit crimes, AIs tend to pinpoint BIPoC people more often than white people.
The potential to change how humanities research is done
In the coming years, what do you expect will be the most significant effect that AI has on the humanities?
I think AI has the potential to change how humanities research is done. Something that was discussed at the conference is that AI is much better at illuminating past patterns than it is at predicting the future. Therefore, humanists can use this understanding of the past to imagine a better future.
I also think AI will give the opportunity for people to access and interact with humanities topics in new ways outside of the typical lectures or museum exhibits.
What challenges does AI present for the humanities, and what opportunities does it present?
A challenge for the humanities is keeping up with the rapid pace of AI and technology in general. I think AI gives the humanities a unique opportunity to re-examine what it means to be human.
To learn more about the intersection between artificial intelligence and the humanities, check out the recent Data & Democracy series presented by the Center for the Humanities and the Public Sphere.