Essay Finalist … Shawn Ray
One of our top five essay finalists was Shawn Ray from Bridgeland High School in Cypress, Texas. He and his essay impressed us for several reasons: his overall committment ethics and technology, his ability to build technology and AI focused solutions, his unwaving determination to uncover issues of biases and limitations of technology and AI, as well as his forward-thinking mindset.
Shawn will undoubtedly continue to have an incredible impact on technology and eithics. We invite you to read his essay below.
Essay from Shawn Ray:
The most pressing challenge I face as a student technologist lies in the unintentional exclusion baked into adaptive learning systems. This realization crystallized during my 900+ hours tutoring peers through Coding With Shawn, where I watched brilliant students from our ESL program struggle against algorithms that misinterpreted their conceptual understanding as fundamental knowledge gaps. At Mathnasium last fall, I witnessed this pattern repeat when our diagnostic software persistently routed a Vietnamese transfer student through basic arithmetic modules despite her advanced calculus capabilities - the system couldn't parse her unique syntax patterns in word problems.
This systemic blindness extends beyond language barriers. Through Compudopt's computer distribution initiative, I documented how 62% of recipients from Title I schools received generic math software ill-suited for their actual STEM capabilities, perpetuating achievement gaps through algorithmic assumptions about socioeconomic status. My work on StutterOn exposed parallel flaws in speech recognition AI, where standard voice interfaces failed spectacularly for peers with speech differences during our school's debate tournaments last semester.
I'm combatting these issues through three parallel efforts: technical mitigation, community education, and institutional advocacy. Using techniques from my quantum computing research at UT Austin, I'm prototyping data embedding models that capture student capability spectra beyond binary right/wrong metrics. My Code Next LMS project lets teachers create culturally responsive test cases - we recently implemented Navajo-based word problems for our Native American Student Association's coding club. Through FBLA, I'm organizing "Algorithmic Bias Hackathons" where students audit educational software using my GitAnalyzer toolkit.
Educators can amplify this work by implementing "Ethical Tech Labs" - I've prototyped such modules using my matplotlib-visualizer to demonstrate how algorithmic decisions get made. Schools should mandate transparency reports from edtech vendors, a concept I'm helping draft through Stanford's AI ethics consortium. Organizations like SEUT could create student advisory boards to pressure-test new technologies, modeled after the Indigenous AI governance frameworks I published last year.
Ultimately, my dual perspective as both developer and student activist positions me uniquely to bridge this gap. Every line of code I write for StutterOn's therapy algorithms gets vetted by our school's speech pathology club, ensuring our AI respects neurological diversity. This iterative, community-driven approach - blending technical rigor with human insight - forms the blueprint I'll bring to reforming educational technology at scale.
