Search papers, labs, and topics across Lattice.
This paper introduces a formal model to differentiate and quantify Agentic Technical Debt (accumulated design liability) and Stochastic Tax (recurring operational burden) in AI agent workflows. The model is expressed as a dashboard equation, expanded into a structural model with defined variables, and demonstrates how to estimate cost categories from operational data. A simulation of accounts-payable processing illustrates the framework's application and provides a companion spreadsheet.
Untangling "Agentic Technical Debt" from "Stochastic Tax" reveals hidden costs in AI agent workflows, offering a clearer path to ROI.
Agentic AI systems combine probabilistic reasoning with delegated action through tools, context, memory, orchestration, and external workflow integration. This note develops a formal and managerially usable model that distinguishes Agentic Technical Debt from Stochastic Tax. Agentic Technical Debt is a stock of accumulated design and governance liability. Stochastic Tax is a recurring flow of operating burden that arises when stochastic agents are used in business workflows. The two constructs are related, but they are not the same: debt can amplify the tax, while the tax can remain positive even when debt is minimized. The note starts from a compact dashboard expression, expands it into a fuller structural model, defines all variables and parameters, shows how each cost category can be estimated from operational data, and illustrates the framework with an accounts-payable simulation and companion spreadsheet.
