UvA-led consortium wins long-term NWO support for trustworthy AI
Artificial intelligence has the potential to help bring about sustainable solutions to many of society’s most intractable problems, and in doing so, to create great opportunities for the Netherlands. But to ensure that AI technology is widely adopted and benefits all sections of society, it is essential that the systems that we develop are trustworthy. The ROBUST long-term programme was set up to tackle this challenge.
Essential hallmarks of trustworthy AI
The ROBUST programme, which has a total budget of over €95 million, will build on the Innovation Center for Artificial Intelligence (ICAI). It intends to add 17 labs to ICAI’s current ecosystem of 30 labs, in areas as diverse as health, energy, logistics, and services. The labs that make up ROBUST will be marked by the economic opportunities they create and their contributions to the UN’s sustainable development goals. They will develop AI-algorithms that advance the state of the art in accuracy, reliability, repeatability, resilience and safety – all essential hallmarks of trustworthy AI.
The project leader is Maarten de Rijke of the University of Amsterdam and ICAI. The co-applicants are Mark van den Brand (Technical University Eindhoven), Arie van Deursen (Delft University of Technology), Bram van Ginneken (RadboudUMC), Eva van Rikxoort (Thirona), Clarisa Sánchez Gutiérrez (University of Amsterdam), and Nava Tintarev (Maastricht University).
CWI spin-off company DuckDB Labs helped create startup MotherDuck which aims to connect DuckDB to the cloud. MotherDuck sports some big names: its CEO is Jordan Tigani, founding engineer at Google’s BigQuery, Google’s fully managed data analysis platform. A big part of the $47.5 million funding comes from Andreessen Horowitz, a prominent venture capital firm, specialized in technology startups.
Breaking ground as the first international conference on Hybrid Human Artificial Intelligence, HHAI22 held its first-ever in-person meeting in Amsterdam in the summer of 2022, establishing the beginnings of an international research community. ADS contributed to the conference by hosting a Meetup around the topic of Hybrid Intelligence.
There are legal rules and ethical frameworks, but little or no practical guidance on responsible design. In her inaugural lecture “System error, please restart”, Nanda Piersma argues what ‘responsible’ means and how we can carry out the (further) development of IT systems in such a way that it also earns our trust.