top of page

About

Marina Shallcross

Welcome! Thank you for taking a look at my work. I am a current graduate student at the University of North Carolina at Chapel Hill pursuing an M.A. in Educational Innovation, Technology, and Entrepreneurship (MEITE). While in the MEITE program, I took courses in learning sciences, human-centered design, innovation and prototyping, learning analytics, statistics using R, and education policy and research.

​

I received a Bachelor of Arts in English Language and Literature from Davidson College in North Carolina in May 2020 and spent the past two years working at EAB, an education technology and research firm based in Washington, D.C., on their K-12 research marketing team. 

​

While at Davidson, I took multiple writing courses that explored the impact of narrative development on memory, agency, and community-building, as well as courses in the digital humanities to learn about and create platforms for digital storytelling. I wrote an honors thesis my Senior year that explored perceptions of memory and identity through my relationship with my autistic brother, juxtaposed with research about intellectual and physical disabilities, the politics of representation, and excerpts from the autism community.

​

At the heart of my interest in the humanities is the idea that writing and the articulation of ideas is a form of agency. Understanding that there are multitudinous barriers for access to quality, personalized reading and writing instruction, and that writing might not be an accessible form of expression and communication for different communities, I became interested in the ways in which we can use technology, specifically artificial intelligence, to help facilitate expression and communication in an increasingly connected society.

 

My interest in artificial intelligence

Since I can remember, I've been interested in science fiction and the speculative nature of the plots and characters, and I feel a similar way now learning about AI and a future where humanity is so deeply interwoven with technology. AI is by no means fictitious, but I was enthralled by the discussions about the future implications of widespread development and dissemination of AI , and a lot of those conversations reminded me of science fiction stories I've read. One of my favorite authors is Ted Chiang, who writes science fiction that explores topics like free will, quantum mechanics, and AI sentience and ethics. What makes his writing resonate with me is that many of his stories share a common reality - ours. Nothing feels too far off from what we are experiencing now.

​

One of his short stories, "The Lifecycle of Software Objects" explores the creation of AI animal-robot companions for humans that develop agency but are commodified and exploited (Chiang, 2010). They experience life and start to grow and change from these experiences. They develop complex emotions and tools for self-regulation, but they remain at the whim of human control. Many ultimately suffer mistreatment at the hands of their owners, while other owners foster an attachment to the companions and have difficulty rationalizing the robots' insentient state (Chiang, 2010).

​

So much of being human is being able to change, to prove to ourselves and others that we are dynamic and that we grow from our experiences, that they mean something more to us than just the moment in which they existed. I think that the distinctions between human and technology will become muddled as AI continues to develop learning principles that allow it to grow and change and start to garner tools for reflection, something that is so human, but can be replicated by technology.

​

We're a long way from this point, and it will not come without serious ethical considerations and regulations. But there is a future not too far away with AI that is so developed that it blurs the lines of sentience. I think about this a lot in terms of how future generations will learn and socialize. Will we have AI teachers? Will AI tutors become more than just an anonymous chatbot? 

​

I think that the world that our children and our children's children will live in will look immensely different than the world we grew up in and live in now. Their whole educational experience will be so much more personalized and efficient, but there are also ways in which it will be worse in terms of fostering attention and thoughtfulness; I want to be on the side that uses generative AI to enhance education and promote critical thinking and a love for learning. I want students to be harnessed with a holistic understanding of what AI is capable of, and what it means to be human and engage with real ideas.

​

​

​

About my project topic

There is so much change and growth in the field of artificial intelligence currently. Right when I learned about ChatGPT and used it for the first time in December 2022, my head started spinning, and I was filled with a sense of alarm. Something about it did not sit right with me. I felt as if so many human capabilities and processes were reduced to a simple prompt-and-response model. I feared that writing would become a lost art, that we would all become complacent and not challenge ourselves to create anything, because we would always feel as if technology could do it better than we ever could. These thoughts were quite alarmist, but not unfounded. We all had, and still have, a lot to process with the AI revolution, especially in terms of how it will impact education, but it does not feel as daunting as it did a few months ago. 

​

I chose this project topic because I wanted to learn more about AI in education and try to make sense of what our world will look like five or ten years from now if we continue at this rate of rapid innovation. The more that I learned, the more positively I began to view AI. Companies are already personalizing learning to a level we have never seen before and giving students a safe space to ask questions and make mistakes. This is amazing! I would have loved to have this as a shy student who never raised her hand in elementary school. I would have answers to, or at least guidance for, all of my unasked questions. I also remember having a lot of ideas sitting in my head that never came to fruition. With chatbots, you can experiment with ideas and build upon them with the help of AI. Asking for and receiving feedback has never been easier. It might not be at the level of teacher feedback, but it is better than nothing. And this is just the beginning.

​

I know that the future of AI will bring about a plethora of ethical concerns and considerations, spanning from how to address biases to how to regulate its development and ownership and how to classify its human-like capabilities (as mentioned in the previous section). A lot of these conversations have already been taking place without producing specific answers. For now, we are in the time of rapid creation of AI tools to make learning more efficient and effective.

click to read the story
bottom of page