Safer AI for a better world

Designing Machine Intelligence for

Cyber Security, Machine Vision, and Data-driven Business Applications


Read more about the next editions

Half-day Tutorial on Adversarial Machine Learning

Attack Prophecy

Rewriting the rules of protection

AI-based Web Services Protection Solution


Trustworthy AI

Pluribus One is a research-intensive startup company that turns basic research results into commercial products and provides innovative AI based solutions and Secure Machine Learning products that are the strongest link in the Cybersecurity chain and not the weakest one.

Adversarial ML Pioneers

Pluribus One is a spin out of the PRA University Lab with more than 20 years of experience in world-class research on Adversarial Machine Learning and in providing solutions based on Pattern Recognition and secure Machine Learning technologies for real-world applications.

Secure your business

Pluribus One develops customized solutions and other data-driven applications to secure your business and your devices. Our customers safely enjoy our products in these fields: Public Administration, Healthcare, Finance, Defence, Education, E-commerce.
Attack Prophecy® is the most advanced system for the detection and protection against web attacks, output of years of research in the field of computer security.
Attack Prophecy® works in three steps:
(1) learning, (2) detection and (3) protection.
It automatically learns the legitimate (normal) traffic profile by observing its live traces.

Artificial Intelligence behind your applications

Machine Learning behind your protection

Discover the effectiveness of our Next Generation
Web Services Protection Solution

Want to see a live demo?

Schedule a Trial

No software agent required.
Simply safe, from any device

Privacy compliant and
tremendously scalable

Want to be involved in the AIsafe DNS Beta-testing?

AIsafe DNS is a comprehensive solution for the prevention and detection of endpoints threats.

It offers coverage against a wide range of threats, from malware to phishing, enabling the mitigation of the risk associated with them.
AIsafe DNS allows to promptly identify machines in a network which contacted malicious hosts on the Internet, so that it is possibile to isolate them immediately. This prevents an attack campaign to result effective on a large scale significantly reducing the damage it may cause.

Additionally, the solution allows to maintain the monitored network healty and clean, enabling the discovery of unwanted or misconfigured services on the monitored endpoints.

Our Partners

Leading Research

The founders of Pluribus One have a large experience in leading R&D projects in computer security, pattern recognition, machine learning and related research areas, funded under the European Research Framework Programmes (FP6 - FP7 - Horizon2020)
Funded by EU under the H2020 Programme.
SIMARGLE's main goal is to combat the pressing problem of malware. It aims to tackle the new challenges in the cybersecurity field, including information hiding methods, network anomalies, stegomalware, ransomware and mobile malware.
Funded by EU under the H2020 Programme.
The main goal of ALOHA is to facilitate implementation of Deep Learning algorithms on low-power embedded systems and heterogeneous low-energy computing platforms, providing automation for optimal algorithm selection, resource allocation and deployment.
Funded by EU under the H2020 Programme.
LETSCROWD overcomes challenges preventing the implementation of the European Security Model with regards to mass gatherings, providing assessment methodologies and advanced automated softwares to Law Enforcement Agencies  and security operators.



Steganalysis and Machine Learning: a European answer
An article from our blog. Steganography is a secret mechanism for encoding information by any means of transmission. Its use has been known since ancient Greece and defined in the glossaries towards the end of the fifteenth century. Both encoding and medium of transmission are secret, that is, known only to the parties who intend to communicate in an occult way.
Steganography differs from cryptography, in which encoding of information and medium of transmission are generally known (think for example to the HTTPS protocol used by this site)
Machine learning may represent a sophisticated weapon at the service of those who intend to unmask steganography. ...


Secure ML Demo - Deep Learning Security
This web demo allows the user to evaluate the security level of a neural network against worst-case input perturbation. Adding this specifically designed perturbation is used by attackers to create adversarial examples and perform evasion attacks by feeding them to the network causing it to fail the classification. In order to defend a system we first need to evaluate the effectiveness of the attacks. During the security evaluation process, the network is tested against increasing levels of perturbation, and its accuracy is tracked down in order to create a security evaluation curve. This curve, showing the drop in accuracy with respect to the maximum perturbation allowed for the input, can be directly used by the model designer to compare different networks and countermeasures.


WILD Patterns Tutorial on Adversarial Machine Learning

WILD PATTERNS is our successful free tutorial on the security of Machine Learning and Artificial Intelligence systems, that has seen over the last years the overall participation of more than 1000 attendees around the world. WILD PATTERNS introduces the fundamentals of adversarial machine learning, presenting recently-proposed techniques to assess the vulnerability of machine-learning algorithms to adversarial attacks, and some of the most effective countermeasures. We consider these threats in different application domains (object recognition in images, biometric identity recognition, spam and malware detection). Don't miss the next edition of WILD PATTERNS, it will be announced soon. The last edition was held at MLS 2019, September 9, Padua, Italy.

Our products are partially developed with the support of Regione Autonoma della Sardegna
(POR FESR RAS 2014-2020 - Asse 1 Azione 1.1.3)


Pluribus One S.r.l.

Via Bellini 9, 09128, Cagliari (CA)


PEC: pluribus-one[at]


Legal entity

Share capital: € 10008

VAT no.: 03621820921

R.E.A.: Cagliari 285352


University of Cagliari

  Pluribus One is a spin-off

  of the Department of

  Electrical and Electronic Engineering

  University of Cagliari, Italy


© 2020 Pluribus One s.r.l. All Rights Reserved.