Establishment of a Working Group in Adversarial AI and Adversarial Machine Learning

Denmark is widely seen as one of the most digitized countries in the world, managing huge amounts of data in digital form. Ongoing efforts to improve welfare among all Danish residents are increasingly based on artificial intelligence and machine learning techniques with the goal to provide algorithmic solutions for future decision-making challenges of the State. In sharp contrast, artificial intelligence systems are known to be sensitive to small errors in the input data, making them vulnerable to attack, intentional misconfiguration, and misuse. This can have detrimental effects on society, threatening the ambition to maintain equality and trust within Danish society.

We propose to form a working group at the Center for Information Security and Trust tasked with
1.Studying the challenges and risks of algorithm decision-making, adversarial AI, and adversarial machine learning on Danish society
2.Researching the robustness of AI and machine learning in general terms
3.Developing recommendations for Danish stakeholders

To form this working group, we propose to hire one person with a research profile for 6 months to
1.Identify and approach stakeholders in Denmark, including universities, businesses, and public institutions
2.Arrange meetings and workshops
3.Arrange all internal and external communications
4.Draft a report describing initial findings to be presented in December 2020
5.Assist with identifying partners and preparing a grant proposal to secure future funding for the working group.