The Primacy of Control: A Risk-Centric TCO Framework for Generative AI and the Financial Irrelevance of Productivity

Generative AI is framed as a productivity enhancer in the prevailing narrative. This paper challenges this view as financially incomplete and potentially misleading for strategic investment. A new framework for financial evaluation was proposed, based on control and risk. Application Programming Interfaces (APIs) and fine-tuned Open-Source (OS) techniques were compared using a stochastic Total Cost of Ownership (TCO) framework that we developed and tested. To deconstruct the key drivers of financial performance, the model incorporates probabilistic estimates for operational and risk variables. These estimates were then analyzed using Monte Carlo simulation and Sobol sensitivity analysis, and the classical value logic of Information Technology (IT) was fundamentally inverted. Sensitivity analysis demonstrates that traditional productivity gains are financially irrelevant in determining the optimal strategy. The model's result was mostly determined by the critical error probability and cost (λ, P). The simulation revealed that the OS strategy has a low likelihood of being financially superior (۶.۵۸%) due to the high cost of its insourced risk management (the HIL process), which significantly affects its total cost of ownership (TCO).

کلیدواژه ها:

Generative AI ، Investment Appraisal ، Risk Management ، technology strategy ، Total Cost of Ownership (TCO)