top of page
Research
Exploring the technical nuances of how AI architecture influences system stability, operational logic, and decision-making within enterprise environments.
Publication 1
When AI Outputs Gain Authority: Model-Mediated State Transitions in Enterprise Frameworks
This theoretical study identifies the shift from fluid generative outputs to fixed procedural authority. It explores the risks of systems processing semantic data as rigid triggers for critical business state transitions and logic.
James Myhre - Working Paper - September 2026 - Zenodo
Publication 2
From Logic Errors to System Failures: A Novel Simulation Method for Assessing Error Cascades Throughout Complex Agentic Architectures
Introducing a simulation method to track how inaccuracies migrate from agent prompts through middleware into final system states. This research maps specific architectural traits that either mitigate or accelerate failure cascades across layers.
James Myhre - Working Paper - September 2026 - Zenodo
bottom of page