We're launching GT as a new research wing focused on distributed decision-making architectures starting Summer Semester 2026.
Mathematically-guaranteed distributed & trustless decision-making only first became possible in the advent of blockchain distributed computing. Figuring out how to make these systems scalable, secure, accurate and cost-efficient is no trivial task. Our mission is to provide mathematical models, algorithms, and deployable architectures for delegative voting systems built on artificial intelligence and smart contracts.
Constrained Optimization of AI Delegative Systems
Our TBC x IF Lab collaboration is focused on solving a major issue with modern ERC-20 standards. In many token-based governance implementations, delegates cannot easily split voting power across competing choices within a single governance decision. As a result, accurately representing a large, diverse electorate can require creating many bespoke delegates—potentially one per voter preference profile. As governance shifts toward AI-controlled smart-contract delegation, that design can become expensive: on-chain actions (delegation updates, vote casting, accounting) risk scaling with the number of tokenholders. We are addressing this by designing an intelligent voting/assignment algorithm that lets a smaller set of delegates approximate the electorate's preferences under formal constraints, while minimizing gas costs.
Team Members
Partner note: routed via Industry/External Affairs