On February 26, 2025, around 300 researchers, industry professionals, and decision makers gathered at Johannes Kepler University Linz (JKU) to discuss the project’s vision and goals. Two Vice Rectors of TU Wien, Peter Ertl (Vice Rector Research, Innovation, International Affairs) and Wolfgang Kastner (Vice Rector Digitalization and Infrastructure), attended the event, as did Gerti Kappel, Dean of TU Wien Informatics.

In the FWF-funded Cluster of Excellence “Bilateral AI”, scientists from Austria’s leading research institutions JKU Linz, TU Wien, WU Wien, TU Graz, AAU, and ISTA, are working together to create a “Broad AI” with advanced problem-solving abilities. Headed by Sepp Hochreiter (JKU), the team consists of 46 key researchers. VCLA Co-Chair Agata Ciabattoni and VCLA Board Member Thomas Eiter are part of the cluster’s Board of Directors, and their TU Wien colleagues Kees van Berkel, Robert Ganian, Georg Gottlob, Thomas Lukasiewicz, Silvia Miksch, Nysret Musliu, Magdalena Ortiz, Mantas Šimkus, Emanuel Sallinger, Stefan Szeider, and Stefan Woltran are also part of the team. To support the project and train the next generation of AI researchers, the participating institutions are actively recruiting around 50 PhDs and postdocs (click here to learn about open positions). The cluster will operate for five years with a funding of EUR 33 million, with the option of an extension of additional five years (increasing the funding to EUR 70 million).

The project aims to create the next level of AI through a combination of symbolic AI (which works with clearly defined logical rules) and sub-symbolic AI (aka machine learning). This combination will result in “Broad AI”, which will be far more comprehensive and flexible than earlier versions, and will thus be able to abstract knowledge and apply it to complex problems. The envisioned “Broad AI” could, for example, intelligently regulate power grids, optimize climate models, and make healthcare processes more reliable and efficient. The abilities of “Broad AI” will not be limited to data processing, but will encompass a profound understanding of identifying relationships and interactions. Combining symbolic and sub-symbolic AI can lead to incredible results, as can be glimpsed from AlphaFold by Google Deepmind, a scientific breakthrough that has been recognized with the 2024 Nobel Prize in Chemistry.

Back row, from left to right: Christoph Lampert (ISTA), Thomas Eiter (TU Wien), Gerhard Friedrich (AAU)
Front row, from left to right: Axel Polleres (WU Wien), Martina Seidl (JKU), Sepp Hochreiter (JKU), Agata Ciabattoni (TU Wien)

Below are two quotes by Sepp Hochreiter and Agata Ciabattoni explaining the aims of “Bilateral AI”:

Univ.-Prof. Dr. Sepp Hochreiter (JKU, Leader of the Cluster):

“AI systems are already an integral part of our daily lives. They are ideally suited to their specific tasks, but once the environment undergoes slight changes, or non-standard situations arise, their problem-solving capacity is severely affected. Examples can be found in medicine, climate change, or energy supply. Our vision of ‘Broad AI’ is more general and works also in the event of changes or exceptions, so this is a completely new level of artificial intelligence. To achieve this vision, we combine the entire expertise of Austrian AI research. We have a very ambitious goal in the interest of the people and for the benefit of our society.”

Univ.-Prof. Dr. Agata Ciabbatoni (TU Wien, Member of the Board of Directors and Deputy Director of Research Module 7, Ethical AI Systems):

“We strive towards creating an AI that is capable of rational thinking, and acts in a way that is legally sound, ethically sensitive and socially acceptable. An important pillar of the cluster is the promotion and training of young scientists in both areas of AI—machine learning and logic-based AI—the fundamental building blocks of a so-called ‘Broad AI.’ In shaping the next generation of top AI researchers, we place particular importance on supporting female doctoral students.”

For more information on “Bilateral AI”, please visit: https://www.bilateral-ai.net/home

This article is based on the information provided at: KI made in Austria: FWF Cluster “Bilateral AI” unter der Leitung der JKU nimmt Arbeit auf