The KRINO project

Researchers and valorization partners of the LANCAR group are involved in developing KRINO, a glass box AI engine designed to enable the user to analyze, evaluate, and produce argumentative discourse.

The software can be used for debunking fake news, checking the trustworthiness of information, and as an aid for decision-making. KRINO will also be able to provide suggestions for improving the quality of argumentative texts produced in a great variety of domains, such as opinion pieces, legal pleas, and student papers.

KRINO’s theoretical framework is based on cutting-edge research in linguistics, argumentation, and causality. The software transforms natural language into formal linguistic structures, identifies the types of arguments occurring in the text, and evaluates them based on an informational account of causality. An important characteristic of KRINO is that it does not function as a “ministry of truth” but is entirely developed in the service of human beings. The software communicates in natural language and is aimed at helping people to understand and assess specific claims or pieces of information.

Since KRINO is meant to be an explainable AI, a lot of attention is paid to transparency, explainability, and compliance with ethical guidelines, both by developers and users. The software is open source and can be consulted on GitHub.

In November 2019, the KRINO project made it to the finals of the Amsterdam Science and Innovation Awards (AMSIA), involving the making of a short promotion video.

Jan van Oort, the initiator of the project, sadly passed away in May 2020. An obituary was published on the webpage of the company he was working for. For the first steps in the development of KRINO, the project is also indebted to Peter Lieber, Roman Bretz, Stephen Hansen, Federico Gobbo, and Marco Benini.

The KRINO team

At present, the code is continuously updated and further developed by Ondrej Uzovic, the project management is in the hands of Hannah Schindelwig, and the team of researchers consists of Davide Ceolin, Federica Russo, and Jean Wagemans.

Relevant publications

Brave, R., Russo, F., Uzovic, O., & Wagemans, J.H.M. (forthcoming). Can an AI analyze arguments? Argument-checking and the challenges of online information quality. In Ch. El Morr (Ed.), AI and Society: Opportunities and Tensions. Taylor and Francis. 

Hinton, M., & Wagemans, J.H.M. (2022). Evaluating reasoning in natural arguments: A procedural approach. Argumentation, 36, 61-84. DOI: 

Wagemans, J.H.M. (2021). Argument Type Identification Procedure (ATIP) – Version 4. Published online December 30, 2021. URL =

Gobbo, F., Benini, M., & Wagemans, J.H.M. (2019). Annotation with Adpositional Argumentation: Guidelines for building a Gold Standard Corpus of argumentative discourse. Intelligenza Artificiale, 13(2), 155-172. 

Wagemans, J.H.M. (2019). Four basic argument forms. Research in Language, 17(1), 57-69. DOI:

Ethical guidelines