Six months into building Tedio (October 2024), I hit a wall. Parents didn't see the urgency of the attention economy's impact, and without immediate results, they weren't willing to pay. Meanwhile, changing Big Tech's algorithmic incentives felt impossible without policy backing.
Instead of forcing a failing model, I reframed the problem: If the system wouldn't change from within, could policy force a shift? This question launched my journey from product development into AI ethics research and policy advocacy.
The Pivotal Questions
The Reframe
When market forces alone couldn't drive ethical change, I realized the need for systemic intervention. Three critical questions emerged that would guide my research direction:
1. What regulations already address algorithmic harm?
Understanding the current legal and policy landscape for algorithm accountability.
2. Who's pushing for change, and what's the momentum?
Mapping the ecosystem of advocates, organizations, and initiatives driving reform.
3. What critical gaps exist in Big Tech's trust and safety efforts?
Identifying where current industry self-regulation falls short.
Global Stakeholder Research
I systematically reached out to key organizations and experts, focusing on understanding the policy landscape and identifying gaps in current approaches:
Policy Leaders & Institutions
HHS (US Dept. Health & Human Services)
Federal health policy and children's welfare
LSE Digital Futures for Children
Academic research on children's digital rights
UChicago Harris School
Policy research and validation
EU Commission
European AI governance and regulation
Legal & Advocacy Organizations
Tech Justice Law Project
Legal advocacy for tech accountability
Alannah & Madeline Foundation
Child-focused AI policy advocacy in Australia
Industry & Standards
YouTube Trust & Safety
Content moderation and child safety
Common Sense Media
Child media research and advocacy
Dual-Pathway Approach
This discovery reshaped Tedio's strategy into a comprehensive dual approach:
🏛️ Policy Influence Track
Working with government agencies and advocacy groups to embed VSD frameworks into child-safety standards and regulatory requirements.
Collaborating with Tech Justice Law Project on legal database systems
Contributing to policy recommendations through stakeholder analysis
Translating technical research into regulatory implications
🔬 Empirical Validation Track
Conducting rigorous research with parents and educators to prove that developmental psychology-informed algorithms can replace addictive engagement loops.
Value Taxonomy Analysis with LO*OP Center (181 academic papers curated)
Interviews with 8 VSD researchers on critical pain points
Collaboration with Seoul National University's HOLI Lab
Academic Research Journey
Thinking in Infinite Decimals
"The universe exists in infinite decimals; there is no 'always' or 'never.' Yet here we are, trying to replicate this boundless reality with 0s and 1s. How can we encode the nuanced spectrum of human values using only binary?"
- intro of my Stanford transfer application essay (*spoiler* I didn't get in :> )
Tedio Internal: Parent Stakeholder Research (2024-Present)
Understanding the "Intentionality Gap" in Digital Parenting
Key Findings:
78% of parents felt unsure about media safety
68% lacked effective developmental tools
Parents want control over algorithms, not just content filtering
Identified "intentionality gap" - algorithms decide what children see instead of parents
Source: Tedio user research documentation (22 interviews, 46 surveys) and Sorenson grant application materials
Seoul National University HOLI Lab (2025-Present)
Research Intern w Prof Yohan Jo at Human Oriented Language Intelligence Lab
Investigating "Value-Based Engineering" Spiekermann: how empirically validated framework can address the "infinite decimals" challenge of encoding human values.
Collaborating with Prof. Zicari (Z-Inspection founder) on ethical AI assessment frameworks. Researching adaptive systems that embrace uncertainty rather than seeking fixed moral truths.
Tech Justice Law Project (2025-Present)
Translating Technical Research into Policy Recommendations
Created systematic approach for converting AI research into regulatory language by designing an internal database
Developed stakeholder and bias analysis frameworks
Built database system for visualizing research insights for policymakers
LO*OP Center Value Taxonomy Analysis (2025)
Identifying Critical Gaps in VSD Practices
Interviewed 8 Value Injection LLM and VSD researchers
Curated 181 academic papers on value taxonomies
Designed framework to enhance VSD methodologies
Identified static value definitions as key limitation in current VSD frameworks
Northwestern Feinberg School Research (2024-2025)
Value-Sensitive Algorithm Impact Study (Designed - Implementation Paused)
Research Design: Between-subjects experimental study investigating how value-sensitive algorithms impact children's identity formation and self-expression compared to commercial platforms.
Status: Study methodology completed but implementation paused due to circumstances.
"Real progress must be more than a technology sandbox that destroys the tried and tested for a little bit of efficiency and comfort. It should be about bringing positive human values to fruition." - Sarah Spiekermann
Key Research Contributions
Parent-Led Design Insights: Documented the "intentionality gap" in current parental control solutions
VSD Implementation Analysis: Identified critical pain points in value-sensitive design practices through expert interviews
Policy Translation Methodology: Created frameworks for converting technical AI research into regulatory recommendations
Interdisciplinary Bridge Building: Connected computer science research with philosophy, law, and child development through multi-institutional collaborations
Vision: Paradigm Shift in AI Ethics
My goal is to contribute to a fundamental paradigm shift in media algorithms, especially for children, by:
Developing Technical Frameworks: Creating concrete implementations of VSD that replace engagement-based models
Advocating for Regulatory Change: Working with policymakers to embed ethical considerations into legal requirements
Building Academic Consensus: Contributing to scholarly discourse on responsible AI development
Demonstrating Viability: Proving that ethical algorithms can be both technically feasible and commercially sustainable