About Mathias Plus
Building the Future of Unsupervised AI Governance
Mathias Plus is a French deep-tech software company developing technologies for Unsupervised AI Governance.
We believe the next frontier of AI governance is not only governing recommendations produced by AI systems, but also governing the structures, segmentations and relationships those systems discover before decisions are made.
Through Deterministic Structural Intelligence, Mathias Plus provides the technological foundations required for stability, explainability, reproducibility and auditability in AI-driven decision processes.
Our Vision
Organizations increasingly rely on AI systems to identify patterns, segment populations, detect anomalies and uncover hidden structures within complex datasets.
As these discoveries become part of operational and strategic decision processes, a new challenge emerges.
How do we govern what AI discovers?
We believe this question defines a new category: Unsupervised AI Governance.
Mathias Plus was created to address this challenge by combining mathematical research, deterministic machine learning, software engineering and enterprise-grade deployment.
Research Origins
MathIAs+ originates from several years of research dedicated to addressing long-standing challenges in unsupervised machine learning.
While supervised learning benefits from objective performance measures, clustering remains affected by issues such as random initialization, limited reproducibility, difficulty comparing segmentations and governance concerns in production environments.
These limitations motivated the development of a new deterministic clustering framework designed to provide fully reproducible results, structural continuity across segmentation levels, industrial scalability and improved explainability.
These research efforts evolved into Deterministic Structural Intelligence, a technological foundation designed to support the governance of AI-discovered structures in enterprise environments.
The underlying innovation has been formally filed in France through the INPI.
From Research to Industrial Deployment
MathIAs+ Structural Intelligence Studio transforms these research results into an operational enterprise SaaS platform.
The solution is designed for organizations requiring reproducible analytical processes, explainable segmentation models, transparent machine learning workflows and stronger analytical governance.
Typical applications include customer and market segmentation, risk portfolio analysis, transaction analysis, anomaly investigation, industrial analytics and exploratory data analysis.
The platform enables organizations to explore AI-discovered structures across multiple levels of granularity while preserving the analytical evidence required for review, audit and governance.
Founder
Hippolyte Lazard-Holly is the founder of Mathias Plus and the inventor of the underlying MathIAs+ technology.
He is an alumnus of École Polytechnique (X89), holds a PhD in Mathematics, and has worked in advanced scientific, engineering and defense-related environments.
His work combines mathematical research, software engineering and industrial deployment with the objective of translating scientific innovation into operational analytical tools.
Mathias Plus reflects this conviction: advanced mathematical innovation can support trustworthy, governable and auditable AI-enabled decision systems.
AI Governance by Design
As organizations strengthen AI governance practices, reproducibility, auditability, transparency and human oversight are becoming increasingly important.
Mathias Plus was built around these principles from the beginning through a deterministic approach to structural intelligence and unsupervised learning.
Deterministic execution supports reproducibility. Structural transparency supports explainability. Generated artifacts support auditability. Human expertise remains central to final decision-making.
Reproducibility → Auditability → Trust → Governance
Technology and Product
The Mathias Plus approach is implemented through MathIAs+ Structural Intelligence Studio, a SaaS platform designed for deterministic multi-K clustering and structural analysis of enterprise datasets.
The platform enables organizations to generate structural indicators, clustering evaluations, decision dashboards and analytical artifacts designed for human-supervised decision processes.
MathIAs+ Structural Intelligence Studio is available through Microsoft Marketplace, enabling organizations worldwide to evaluate and deploy these capabilities in a self-service environment.
ESG Commitment
Mathias Plus believes that innovation and responsibility must progress together. We are committed to developing technologies that support governable AI, sound analytical governance and responsible innovation.
Through reproducibility, transparency, auditability and human oversight, we help organizations strengthen confidence in data-driven decision-making and AI-enabled processes.
We seek to align our activities with principles of ethical business conduct, accountability and sustainable long-term value creation.
A French Deep-Tech Company with Global Ambition
Mathias Plus is based in France and builds software for organizations operating in complex, regulated and high-accountability environments.
Our ambition is to contribute to the next generation of enterprise AI governance by making AI-discovered structures stable, explainable, reproducible and auditable.
This is the foundation of our work on Unsupervised AI Governance.
News
September 10, 2026:
random_state = 42. Is this really AI governance ? Episode 3 : Survival.
Read the PDF →
September 4, 2026:
MathIAs+ is now available on Microsoft Marketplace with Japan data residency.
Read the PDF →
September 2, 2026:
random_state = 42. Is this really AI governance ? Episode 2 : Robustness.
Read the PDF →
August 26, 2026:
random_state = 42. Is this really AI governance ? Episode 1 : The random state illusion.
Read the PDF →
August 24, 2026:
Press Release.
Three Challenges in Unsupervised AI. One Deterministic Response.
Explore on prlog.org →
August 21, 2026:
Three Challenges. One Deterministic Response. The Mathias Plus Solution.
Explore on LinkedIn →
August 3, 2026:
Building the Future of Unsupervised AI Governance.
Read the article on LinkedIn →
July 29, 2026:
Deterministic Clustering for Responsible AI.
August–September 2026 Early Adopter Initiative.
Read the article on LinkedIn →
July 24, 2026:
Beyond AI Act Compliance: Reproducibility, Auditability and Trust.
Read the article on LinkedIn →
July 22, 2026:
MathIAs+ Structural Intelligence Studio now available in Microsoft Marketplace.
Read the press release →
July 17, 2026:
Six Months. One R&D Milestone. One SaaS Platform. One Microsoft Marketplace Launch.
Read the article on LinkedIn →
Follow MathIAs+ on LinkedIn