ASU graduating student applies AI, deepfake technology in entomology research

Editor's note: This story is part of a series of profiles of notable fall 2025 graduates.

Kacy Hatfield, a graduating Master of Science student in media arts and sciences from Arizona State University’s School of Arts, Media and Engineering, is Herberger Institute’s Outstanding Graduating Student for fall 2024. She has spent her academic journey exploring the complexities of artificial intelligence and its applications in ethical, biological and computational spheres.

Guided by her advisors, Professor Pavan Turaga, director of the School of Arts, Media and Engineering, and Associate Professor Theodore Pavlic, Hatfield tackled a project involving intricate AI coding and pattern recognition to observe and manipulate the behavior of wasps, a species known for its complex social structures and ability to recognize individual colony members.

“When I first began, I was relatively new to PythonPython is a high-level programming language. , and coding felt like this massive uphill battle,” Hatfield said. “The first few months were almost entirely filled with failure. I would run a block of code, and it would fail. Then I’d tweak it, run it again and fail again. But with every attempt, I was learning a little more about what didn’t work, and eventually things started to click.”

At the heart of her research, Hatfield used deepfake technology to simulate interactions among wasps, studying how the insects would respond to familiar and altered visual cues in their recognition of each other. By swapping markings on the wasps' bodies, she enabled University of Michigan professor and ecologist Elizabeth Tibbetts to test the boundaries of the wasps’ individual recognition abilities and observe their responses in controlled settings. The wasps’ interactions offer a unique lens into understanding the foundations of social recognition — a concept familiar to humans — yet difficult to ethically replicate outside of such specialized research.

Hatfield explained the motivation behind this approach.

“These wasps are fascinating because they behave in ways that are, in some ways, social. They recognize each other’s faces and form hierarchical relationships based on past encounters. By experimenting with deepfakes, we’re seeing just how far we can push this recognition. We’re finding out what makes a face memorable in the animal kingdom.”

Hatfield’s early work with deepfake technology came with its own set of challenges, particularly in ensuring ethical applications within biological research. She was quick to acknowledge this, noting, “Deepfake technology is usually associated with negative impacts, so using it in a positive, controlled way feels really exciting to me. It’s about turning a potentially controversial tool into something constructive and informative, but it takes a lot of trial, error and ethical consideration.”

Reflecting on her journey from novice coder to contributing researcher in the field of artificial intelligence, Hatfield highlights a newfound appreciation for failure as a tool for learning.

“AI research is honestly just a long process of quick failures,” she said. “At the beginning, it might take me a day to find out what went wrong. Now, I can figure it out in seconds, fail fast and try again just as quickly. I’ve learned it’s through these small failures that the biggest breakthroughs happen.”

After overcoming these technical hurdles, Hatfield was able to present her findings at the prestigious Conference on Computer Vision and Pattern Recognition (CVPR), where she shared her work with a broader academic audience. Her research not only pushes the boundaries of AI but also invites a fresh perspective on animal behavior and social recognition — discoveries she hopes to further explore in a PhD program at the School of Arts, Media and Engineering.

For Hatfield, this experience at ASU has become a testament to resilience, curiosity and the remarkable potential of trial and error in scientific discovery.

In the below Q&A, Hatfield offers more insight into her experience conducting research at ASU.

Note: Answers may have been edited lightly for length and/or clarity.

Question: What was your “aha” moment when you realized you wanted to study the field you majored in?

Answer: My aha moment was at a conference called CVPR, where I presented my research on deepfaking wasps. It really opened my eyes to how computer vision can apply to so many fields I’m passionate about. After the conference, I connected with people deeply invested in using AI for positive impact, and that solidified my decision. I knew I wanted to pursue research in computer vision, especially in applications that I am passionate about, like planetary science and ecological research.

Q: What’s something you learned at ASU — in the classroom or otherwise — that surprised you or changed your perspective?

A: One thing I learned is that progress isn’t linear. In research and in the classroom, I found that failures should be celebrated as much as successes. Finding ways that don’t work gives you just as much insight into a problem as finding the right solution. Learning to celebrate failures has been one of my biggest takeaways while studying at ASU and researching.

Q: Why did you choose ASU?

A: I chose ASU because I loved the program I was in for undergrad — media arts and sciences, which was digital culture at the time. It was the perfect fit for studying computer vision through the lens of a creative. I wanted to apply these principles as an imaginative scientist, and ASU’s School of Arts, Media and Engineering programs offer the ideal environment for that. It gave me the freedom to set my own research goals and deliverables, plus amazing resources and advisors. This turned out to be a great choice!

Q: Which professor taught you the most important lesson while at ASU?

A: This is a tough one, but I’d say Professor Cassie Bowman taught me the most about patience and scientific optimism. Working on the Psyche mission in undergrad, I saw firsthand how delays could be a positive thing, allowing us to extend outreach and engage more with the community. She showed me how to appreciate the process, even in moments of waiting.

Q: What’s the best piece of advice you’d give to those still in school?

A: Celebrate your failures. It can be discouraging when things don’t work out, especially after trying again and again. But persistence is key, and any progress is still progress. It’s important to recognize that even setbacks contribute to the bigger picture.

Q: What was your favorite spot on campus, whether for studying, meeting friends or just thinking about life?

A: The second floor of Hayden Library was probably my favorite spot. It has a great view that overlooks the lawn, and I loved watching everything around campus. It felt like the perfect place to work on my undergrad thesis and graduate projects.

Q: What are your plans after graduation?

A: I’ve arranged an internship with NASA Goddard for the spring semester. I’ll be applying similar computer vision techniques from my research here at ASU to detect dust devils in Martian environments, continuing my work in computer vision with a focus on scientific applications.  

Q: If someone gave you $40 million to solve one problem on our planet, what would you tackle?

A: I’d focus on two areas. First, I’d support research for the International Myeloma Foundation. Multiple myeloma is a terminal cancer close to my heart, as my mom has it. Machine learning could make a real difference, and many IMF-funded researchers are already making strides in this. Secondly, I’d invest in AI education workshops for underserved schools, especially to address fears around AI and machine learning. Teaching kids early could demystify the technology and show its positive potential.