Research Projects
The following projects bring together research developed under Society & AI and led by Sai Gattupalli, Ph.D., Principal Scientist at Society & AI. The work ranges from AI that can function without internet access to lifelike learning companions, K-12 educational media, and open science communication for broader public understanding. For research collaborations, contact research@societyandai.org with a brief project proposal, institutional affiliation, and anticipated funding source.
AI Feedback That Works Without Internet
A published IEEE study on how small AI models can run directly on a student's own device and support coding practice even when internet access is limited or unavailable.
AI Learning Companions
A multi-stage AI pipeline for designing, animating, voicing, and reviewing lifelike learning companions for educational tutoring systems.
STEAM Music Videos for K-12 Learning
An open educational media project that turns science and math concepts into accurate, readable songs and videos for classroom and at-home learning.
Society & AI: Public Scholarship for AI and Education
The umbrella science communication project that brings this research together through accessible writing, open resources, and public-facing scholarship on AI, education, and society.
The work of Society & AI proceeds from a foundational conviction: that artificial intelligence is not a neutral instrument but a sociotechnical system that encodes choices about knowledge, power, and human worth. Our research examines those choices — in design, in deployment, and in governance — with particular attention to how they play out in educational contexts where the stakes for equity and human development are highest.
Each project in this portfolio is an applied expression of that broader agenda, grounded in evidence and oriented toward systems that expand human capability rather than substitute for it. The standard we apply is not what a system can do in optimal conditions, but whether it serves the learners and communities most likely to be left behind when the conditions are not optimal.
The research presented here has been conducted without external grant funding. All projects have been supported entirely through the personal investment of the Principal Scientist — a deliberate choice that preserves full independence of inquiry, free from institutional mandates, funding-body constraints, or commercial influence of any kind. As this research program grows in scope and ambition toward broader societal flourishing, we intend to pursue institutional support through competitive grants, collaborative partnerships, and philanthropic investment commensurate with the scale of the questions we are working to answer.
"We are at an inflection point in which the architectures of learning itself are being redesigned by systems that encode assumptions about knowledge, capability, and human worth. The legitimacy of artificial intelligence in education must be measured not by its technical sophistication, but by its fidelity to human flourishing, epistemic equity, and the democratic participation of communities in the design of the systems that will shape their futures."
— Sai Gattupalli, Ph.D. · Founder & Principal Scientist, Society & AI
University of Massachusetts Amherst
Advanced Learning Technologies Lab