From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems
A new survey paper examines the challenges of ensuring determinism in AI systems used within the financial industry. It highlights how modern AI techniques, including deep neural networks and generative AI, introduce nondeterminism due to hardware and architectural factors. The paper analyzes these issues across tabular models, graph networks, and LLM-based workflows, proposing a framework to evaluate audit readiness by linking specific metrics to determinism levels. AI
IMPACT Highlights critical auditability challenges for AI in regulated financial environments, proposing a framework to improve system trustworthiness.