Introduction Traditional AI observability focuses on system-level metrics such as latency, token usage, error rates, and infrastructure health. While valuable, these approaches provide limited insight into the actual behavior of generative models during or after content generation. As AI systems are deployed in high-stakes production environments, a more precise layer of measurement is required. This […]
Introduction As generative AI systems move from experimentation to production environments, organizations face a growing challenge: how to reliably monitor, understand, and govern their behavior at scale. Traditional benchmarks provide static snapshots of performance, while classical observability tools focus primarily on infrastructure metrics such as latency, throughput, and cost. Neither approach fully addresses the dynamic,
NeoMundi is pleased to announce its acceptance into the NVIDIA Inception program. An Already Operational Runtime Governance Infrastructure NeoMundi joins NVIDIA Inception to accelerate the development of ControlTower, our infrastructure for measurement, auditability, and real-time governance of AI systems. As organizations massively deploy AI in production, the key question is no longer whether a model
In the framework of the NeoMundi AI Observatory’s work, we are publishing today a contribution by James Moore, founder of Nova Jema AI Systems and specialist in the governance of artificial intelligence systems under real-world execution conditions. He kindly shared with us a precise reflection on a frequently overlooked subject: once we measure an AI’s
Scientific Contributor / Data Scientist – NeoMundi AI Observatory Most discussions on AI governance start too late. We believe measurement must come first. We have developed ControlTower, a diagnostic, piloting, and traceability instrument designed to continuously observe the risk of AI responses in real-world conditions. ControlTower enables standardized and reproducible measurement campaigns, allows comparison of
Independent Audit Report by Pape Malick DIOP, Data Scientist and Machine Learning Researcher, June 10, 2026. At NeoMundi, we believe that the governance of generative systems cannot rely on a single signal. This is why we commissioned an independent external exploratory audit of our runtime signals: G-Score, ∆G, FLAG, regime, and benchmark accuracy. Pape Malick
NeoMundi Recherche publishes a public, sanitized version of an independent methodological review of its real-time approach to measuring the stability of generative AI. The review is based on a May 2026 comparative cohort of eight anonymized providers. It examines what a real-time signal can and cannot capture, documents the current limits of the method, and
