The NeoMundi AI Observatory is our public research program dedicated to the continuous, transparent and reproducible evaluation of generative AI systems.
We regularly publish thermodynamic cartographies and in-depth observations, including the NeoMundi AI Barometer, which tracks de-identified model profiles over time across multiple dimensions: stability, factual validity, informational density, factual-risk signals, semantic variation and cost efficiency. These publications reveal behavioral trends and silent regime changes that traditional one-shot benchmarks cannot capture.
All our work is built on open methodologies and publicly available data. Generation scripts, scoring protocols and raw datasets are released on GitHub, allowing researchers, developers and the wider AI community to audit, reproduce and build upon our observations.
NeoMundi is pleased to announce its acceptance into the NVIDIA Inception program. Developing an Independent Runtime Measurement Layer 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 […]
Reading note — 15 October 2026. This publication reports a June–July 2026 campaign. Figures describe signals recorded by the measurement chain used at the time, not verified error rates or a general assessment of model reliability. Cross-period comparisons require matching protocols, evaluator versions and coverage; they must not be carried over automatically to a later
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
All AI Barometers Building a baseline To produce this first Barometer, we first established a public baseline. It was built from 3 execution waves × 12 de-identified systems × 4 fixed questions × 100 repetitions, representing 14,400 completed observations. This baseline is the reference for this June 2026 series. Later comparisons require an equivalent measurement
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, an observation and traceability instrument for AI responses under documented execution conditions. ControlTower supports repeated measurement campaigns to compare response variations across models and over time. We evaluate
This publication concerns a 2026 campaign and the measurement configuration used at that time. Its results are not automatically comparable with those of later instrument versions. 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
This publication concerns a 2026 campaign and the measurement configuration used at that time. Its results are not automatically comparable with those of later instrument versions. 1. For AI systems, price, stability and factual accuracy do not always move together We measured 12 major generative AI systems using the same protocol, in order to compare
This publication concerns a 2026 campaign and the measurement configuration used at that time. Its results are not automatically comparable with those of later instrument versions. 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
1. MURMURATION – Research on Collective Runtime Stability Distributed autonomous systems can now coordinate trajectories, distribute tasks, and maintain dynamic formations at scale. But one critical question remains largely unsolved: how can we measure the collective stability of a multi-agent system before a systemic failure in cohesion emerges? Current distributed architectures are designed to optimize:
