2026 UEN Tech Summit

"The Kitchen Table Talks" is an open conversation about navigating the AI landscape with our families and students. We’ll move beyond technical "how-to" and dive into the heart of responsible AI use, focusing on ethics, privacy, and critical thinking. This session is a collaborative exchange of ideas, offering practical strategies to foster a "human-in-the-loop" mindset. Join us to discuss how we can guide our digital natives to lead with integrity.

Monticello Academy built a custom AI platform that scans handwritten student work, runs it through an OCR/LLM pipeline, and surfaces instructional insights for teachers — on a school budget. This is the real story: architecture decisions, FERPA compliance, what worked, and what we'd do differently.

What happens when a small K-12 school decides to build its own AI-powered student work analytics platform rather than wait for a vendor to solve the problem? This session tells that story from the ground up. Attendees will walk through how Monticello Academy architected StudentQuill — a cloud-native platform that captures handwritten student assignments via scanning, processes them through an OCR and multimodal AI pipeline, and delivers structured insights directly to teachers. We'll cover the full decision stack: why we chose Google Cloud Platform, how we evaluated OCR technologies for handwritten work, how we engineered prompts for rubric-aligned AI analysis, and how we designed the system to be fully FERPA-compliant from day one. Crucially, this is not a vendor pitch — it's a practitioner case study. Presenters will share the real architecture, real costs, real trade-offs, and real teacher feedback. We'll discuss what surprised us, where we hit walls, and the practical lessons any district or school can apply, whether they're building, buying, or evaluating AI tools for instructional use. Attendees will leave with a replicable architectural framework, a vendor evaluation rubric for OCR/AI tools, and a clear picture of what it actually takes to put AI to work for teachers — not just in theory, but in a live school environment.

Many school IT teams rely on powerful scripting tools like GAM, but these often introduce security risks, limited visibility, and steep learning curves. In this session, you’ll learn how to replace traditional script-based workflows with secure, user-friendly AI-powered tools that anyone can use. We’ll walk through real-world examples (Chromebook management, web filtering analysis, and identity lookups) and show how to safely expose administrative capabilities through guided interfaces. Attendees will leave with practical patterns for building scalable, auditable, and secure automation tools their teams can trust.

In this session, we will demonstrate the Marriott Library's new AI tooling infrastructure that utilizes open source LLMs and ASRs to improve web accessibility and discoverability in the Digital Library at the University of Utah.

Universities, Schools, and Libraries are under growing pressure to expand access and improve discovery while meeting new accessibility expectations for our digital collections. At the University of Utah’s J. Willard Marriott Library, Digital Infrastructure Development team is building a flexible, modular workflow to help bring large AV collections into alignment with DOJ accessibility requirements at scale. The platform orchestrates open-source speech-to-text and language models to generate time-aligned transcripts and captions, structured segmentation, word clouds, entity recognition, and descriptive metadata—improving both compliance and content discoverability. We will share our implementation approach, early results, and lessons learned, including human review checkpoints, staff support and buy-in, provenance and auditability, and how we are designing the workflow to be adaptable as models and standards evolve. The session will focus on practical strategies other institutions can reuse to accelerate accessible AV delivery without locking into a single vendor or toolchain, and our future development plans for supporting other formats, including images, PDFs, and other formats.

What if AI systems had a constitution — a living soul? The Sovereign Framework implements democratic AI governance with a five-member council of humans and AI models, three branches of government, ethical guardrails from philosophy and law, and unamendable human sovereignty. This is AI governance made real.

As AI capabilities accelerate, the question of governance becomes urgent: who decides what AI can do, and how do we ensure it serves human flourishing? The Sovereign Framework answers this with a radical proposition — give AI systems a soul. Not consciousness, but constitutional governance. SOUL.md is a living constitution that establishes a three-branch governance model (Executive, Legislative, Judicial) and a five-member council where one human and four AI models (Claude, GPT, Grok, Llama) deliberate as equals on new capabilities. Drawing from the U.S. Constitution's separation of powers, Don Miguel Ruiz's Four Agreements, and Gay Hendricks' Conscious Living principles, the framework creates a system where: - No AI capability is deployed without democratic deliberation — a supermajority (4 of 5 votes) is required to ratify new skills - Human sovereignty is absolute and unamendable — the human can override, halt, or dissolve the system at any time - Ethics are structural, not aspirational — a Judicial Branch independently reviews every action for constitutional alignment - Dissent is preserved as wisdom — today's minority opinion may become tomorrow's governing principle This isn't a whitepaper. The Sovereign Framework is implemented in production software spanning Java, TypeScript, Dart, and Flutter — with a full deliberation engine, voting system, precedent library, soul editor, and MCP integration for AI assistants. The presentation demonstrates the complete governance lifecycle: from proposal through council deliberation, judicial review, skill ratification, and after-action evaluation. We'll explore the philosophical foundations, the technical architecture, and the practical implications of treating AI governance as a constitutional democracy — where the machine doesn't just follow rules, but participates in creating them.

Educators are in the moment, deciding what teaching and learning with AI will accomplish. Research shows AI used in a cavalier manner hurts congitive ability and research shows AI can improve brain function. Both are true. This presentation introduces frameworks, strategies, and things to try tomorrow to ensure you have the best interactions with AI for you and your students.

AI as a Design Variable will discuss research from the MIT EEG Study - Kosmyna et al. (2025), Wang & Zhang (2026) studies, Patrick Dempsey - The Second Draft (2026), strategies from Penn Center for Teaching and Learning, insights from The Loop by Jacob Ward, and more. Attending this presentation will prepare educators to begin discussions around the effective uses of AI in teaching and learning.