We predict the most severe attack intentions.
We have been researching and testing security solutions in adversarial environments for the past 20 years to raise the security bar


Battista Biggio (Pluribus One founder) at MIT Tech Review, Paris
It was in a university laboratory that we were born.
In which we adhere rigorously to the scientific approach to research.
As a result, we have been able to focus on what is really needed in the commercially independent information security industry.
That’s why we pioneered Adversarial Machine Learning, the last frontier in the field of information security.
Several European projects and important national companies have used our algorithms.
The ICML Test of Time Award 2022
The prestigious international prize was awarded to Battista Biggio, co-founder of Pluribus One, for the scientific article: “Poisoning Attacks Against Support Vector Machines.”
ICML awards the Test of Time Award to scientific articles whose impact on the scientific community has been the greatest in the ten years following their presentation. A few of the prestigious predecessors include researchers from Amazon Research, Harvard University, Facebook Research, and Google Brain.

Our international projects.
Pluribus One’s founders have years of experience developing and coordinating projects funded under European framework programs (FP6 – FP7 – Horizon2020) in the areas of information security, pattern recognition, and machine learning.

AIDA – Artificial Intelligence and advanced Data Analytics for Law Enforcement Agencies
Seer Box at the service of European law enforcement agencies: Using Artificial Intelligence and Machine Learning, AIDA will provide data analysis platforms based on descriptive and predictive analytics, as well as tools for preventing, detecting, controlling, and combating cybercrime and terrorism. The platform will rely on frontier research, Big Data analysis and Deep Learning techniques.

TESTABLE – TestabiliTy Pattern-driven Web Application Security and Privacy Testing
As part of this project, Pluribus One pours all its expertise in Adversarial Machine Learning into the protection of Web applications and services. TESTABLE proposes novel metrics for quantifying security and privacy risks within the software development cycle. In the project, Pluribus One is responsible for evaluating the performance of Web Application Firewalls.

AssureMOSS – Assurance and certification in secure Multi-party Open Software and Services
The AssureMOSS project addresses the issue of open-source software and supply chains. The majority of software developed today incorporates third-party components without a process for ongoing and secure certification and recertification. By supporting all phases of the continuous software lifecycle (design, development, implementation, evaluation and backup), the project aims to transition from process-based to artifact-based security assessment.