FINAL PORTFOLIO! Kiryn Watson

Reflective Cover Letter 

Dear Write 329 Classroom Community & Dr. Mallette,

Defining Style

To me, style is defined by the deliberate choices a writer makes in order to communicate with a particular audience in a way that they feel is both rhetorically effective and true to their unique voice. As such, there is variance in certain stylistic elements throughout my portfolio, as well as elements of my style that are present throughout each of my projects. My writing style tends to lean academic, which makes sense, as this is the context that I have done most of my writing in. I tend to write somewhat formally, densely, and logically. It seems like I can’t help but try to make whatever piece I’m working on practical. For example, in my video project, the tone of the video changes from a silly skit to a very information-dense, informative piece about halfway through. As a STEM major, I have been repeatedly told that this is what my writing should be like, and clearly the message has sunk in. Despite this, there is also a bit of humor and playfulness that comes through in all of my writing. Even in my AI Use Philosophy, my voice comes through in unusual word-choices and phrasing, i.e. “monstrosity”, “seemingly unknown to man”, “little quirk” (sarcastically). In my audio essay script, this is a bit more apparent. For example, I compare a young Niall Horan’s appearance to that of Ellen Degeneres.

How I Met Course Learning Outcomes:

Throughout the semester, I created a variety of multimodal texts, including a video text, audio text, and social media post. In order to create these, I used several different technologies–Audacity, Da Vinci Resolve, Pixlr editor (an internet-based photo editing software, comparable to photoshop), and created a Tumblr blog to showcase them all together.

At the beginning of the term, I performed primary and secondary research in order to determine how I would use, or refrain from using AI in my writing. The framework that I came up with is guided largely by the ethic of care, both for humans and the environment. It is also guided by my own values of human creativity and learning. These principles are something that I continued to revisit throughout the semester, as I determined whether AI use would be necessary, or beneficial to my work in any capacity. 

Writing my own texts, and evaluating the texts generated by AI allowed  me to gain a more thorough understanding of how language choices impact rhetoric. Throughout the research I conducted to inform my AI use philosophy, this was a big theme. 

When creating and revising each of my projects, I took the experience of my audience into consideration, and made stylistic choices accordingly. For example, in my wildcard project, I kept the text portion of my comic simple, concise, and casual. However, in my AI use philosophy, my sentence structure is more complex, I use more technical language, and the tone is more formal. 

Finally, this entire process has led me to develop a greater awareness of my audience’s needs, as well as what I can do to meet those needs. 

Technology and Style

Over the course of the semester, I abstained from using AI in my writing for the most part. When I did attempt to use AI to revise a few different pieces of my work, I found myself feeling that it did not do justice to certain intentional stylistic elements. For example, I attempted to use Gemini to edit pictures of my own drawings, simply to enhance visual clarity. However, Gemini ended up moving line placement and removing textures from my drawings, resulting in a product that felt very smooth, symmetrical, and soulless. When working on each of my projects, I was striving to create work that felt personal. Based on my experience, using AI as a major collaborator on any of these projects would not have helped me achieve this goal. 

Key Takeaways

When reflecting on the work that I did this semester, I feel that the experience that I have gained from experimenting with my writing outside of the context of my STEM-related pursuits has been extremely valuable. At the beginning of the semester, I felt a bit nervous about being the only non-writing major in this class. Learning to identify what elements of STEM-based writing bleed over into any other writing that I do has helped me become a more intentional writer. This has also given me additional reassurance and confidence that I’m capable of producing successful writing that is not STEM-related. At the beginning of the semester, I felt a bit nervous about being the only non-writing major in this class.

The encouragement to experiment with new projects, tools, and modalities has allowed me to write with my own voice more intentionally, and in a more creative manner than what I am accustomed to. I have had very little multimodal composing experience prior to taking this class, and I’m grateful to have had the opportunity to practice and learn new skills without the pressure to create perfect work instantly. This has also taught me more about what kind of writing I really like to do!

Going forward, I can apply the skills that I’ve developed in this class to any writing that I do in the future. I truly loved being able to work on projects that are a bit more creative than what I have experienced in the past, and am excited to continue doing so. In the context of writing in STEM, I have gained a far more thorough understanding of style than what I have been previously taught. 

Sincerely,

Kiryn Watson

In a recent conversation with my romantic partner, who happens to be an ecologist for USGS, it was mentioned in passing that the majority of his coworkers use GenAI in some capacity to write their publications. I was a bit shocked by this, and when I asked him to elaborate, he said that a lot of the content in the papers that they’re required to produce is “just fluff”. This seemed to be reassuring to him, but it did not ease my mind. As someone who currently produces writing in the biological sciences, it feels a bit alarming that GenAI models are already being adopted by the scientific community in a way that may not be entirely beneficial to the user.

When considering how I should approach AI use in my own writing as a student and a professional of the biological sciences, the ethic of care is at the forefront of my mind. I care greatly about producing work that I feel is of high quality, and I think about the people who might read my work in the future. Right now, for instance, I am working on rewriting an outdated set of instructional laboratory procedures that will be referenced by my current and future coworkers. I feel responsible for making these procedures as easy to use as possible, because I care about the success of my colleagues. With this in mind, it is important to acknowledge that GenAI is not known for consistently producing the highest quality of work, even when used solely in an editing capacity. For example, when I attempted to use Notion AI to revise a brief section of these procedures, some issues came about. I prompted the AI to remove a column from a chart, remove unnecessary procedural steps and list items, and make the document more easily to visually navigate. While these tasks were accomplished, Notion added in some material of its own—supplies, measurements, and other instructions that neither exist in the original text of the procedure, nor were they provided by me. These fabrications of content by AI, referred to as hallucinations, are an inherent result of the vast amount of material that AI models are trained upon (Greengard). To me, this highlights the importance of AI being used exclusively as a tool, particularly one that the user cross-checks. This little quirk doesn’t just cause inconvenience to the user, confusion to the reader, and generate a monstrosity of a product. In a laboratory setting, when working with both hazardous and infectious materials, the unsolicited insertion of nonsense into procedural documents by GenAI becomes an issue of safety when it goes unchecked. When I do choose to use AI in my writing, I will thoroughly review the material that it generates to ensure that I create work that is truly reflective of my intentions.

As someone who highly values learning, skill development, and human creativity, I feel uncomfortable with the use of AI in academic and professional writing beyond the area of revision. Revision, I believe, is something that AI can be quite helpful with, especially when access to another person for this purpose may not be convenient. For example, when I prompted Claude to revise a research proposal that I had written for a behavioral ecology course, provided with the original text, rubric, and feedback from the professor, I was pleased with the work that it did. It altered sections of text that used passive voice that needed to use active voice (as specified by the provided rubric), fixed some formatting and citation errors, and rearranged a few sentences for the sake of clarity. Much to my chagrin, two hallucinations made an appearance, but cross-checking the work of Claude allowed me to identify and remove these. This experience makes me feel comfortable with using AI for this purpose, but I don’t see myself using it beyond this. Researchers at the UDST, a STEM-focused University in Qatar, suggest that the overuse of AI in writing results in a diminished ability of students to think critically and creatively, and hinders their ability to learn necessary writing skills (Charummal et al., 2025). If we become reliant on these tools as a means of production, we deny ourselves the opportunity to learn and grow as writers. As such, my use of AI in writing will be strictly for the purpose of editing and revision, not for the generation of original content.

Being that I currently work in, and plan to continue to work in the field of natural resources, the environmental impact of AI use is another area of concern for me. While experts report that it is difficult to quantify the use of natural resources by AI data centers at this point, comprehensive approximations do exist. The collective annual water footprint of these centers is suggested to be the equivalent of the annual global consumption of bottled water, which is 446 billion liters (de Vries-Gao, 2026). AI data centers are also known for using mass quantities of power. Four percent of total power use in the U.S. in 2024 was attributed to AI data centers, and this number is expected to grow exponentially over the next few years (Leppert 2025). Electronic waste generation by AI data centers is another issue. The chip technology and hardware that AI relies on to run have short lifespans of just a few years. The replacement of these parts generates millions of tons of toxic e-waste, most of which is not disposed of properly and ends up in landfills (Schwaller, 2024). Considering the strain that AI data centers put on our environment, the determination of what constitutes worthwhile AI use is imperative. If, for example, I have to repeatedly alter the prompt that I give to Dall-E to make simple edits to an image, and still can’t get a usable result, then I don’t feel that this use is justified. Essentially, I would like to minimize my participation in a process that is so implicitly wasteful. If an AI tool takes an immense amount of use and reuse to produce a satisfactory result, I will not use it. Moreover, if I can perform the task at hand with a comparable level of efficiency and effort, I will do so instead.

Going forward, I plan to use AI in my writing as a tool for editing and revision that will be thoroughly cross-checked, never as a generator of content. The extent to which I use it will be determined circumstantially by the efficiency of a given tool to perform a particular task, in an effort to limit my contribution to the negative environmental impacts made by AI data centers. As I continue to grow as a writer and biologist, and as advancements in AI technology are inevitably made, the values that shape my current use philosophy will go on to determine my future use.

References

Charummal, R. A., Hamidi, L., Mendes, C., Miranda, M. M., & Hill, C. (2025). Unraveling the dilemma: Examining the adverse effects of AI writing tools on STEM student motivation – Insights from an academic writing center. MENA Writing Studies Journal, 1(1), 74–83.

de Vries-Gao, A. (2026). The carbon and water footprints of data centers and what this could mean for artificial intelligence. Patterns, 7. https://doi.org/10.1016/j.patter.2025.101430

Greengard, S. (2025). Shining a Light on AI Hallucinations. Communications of the ACM, 68(5), 9–11. https://doi-org.libproxy.boisestate.edu/10.1145/3715691

Leppert, R. (2025, October 24). What we know about energy use at U.S. data centers amid the AI Boom. Pew Research Center. https://www.pewresearch.org/short-reads/2025/10/24/what-we-know-about-energy-use-at-us-data-centers-amid-the-ai-boom/

Schwaller, F. (2024, October 28). E-waste from AI computers could ‘escalate beyond control’. DW. https://www.dw.com/en/e-waste-from-ai-computers-could-escalate-beyond-control/a-70619724