New perspectives for cyber security in software development: when End-User Development meets Artificial Intelligence

Current research faces a "war" between Human-Computer Interaction and Artificial Intelligence. Despite the shared goal to support the user in his/her choices, researchers seem to be unable to find a common point between the two topics. While one tends to demonstrate that the user needs to have total control of the system he/she is using, the other aims to create mathematical models that automate as much as possible, to give the user the "feeling" that the system has a brain and can think on its own. In the current work, we want to demonstrate that there is more than one possible meeting point between those two research fields, without diminishing the two ideologies. To do so we propose both the adoption of End-User Development techniques (that is a branch of the Human-Computer Interaction) to collect data, and the adoption of Artificial Intelligence models, to support the user in performing the right choice. Additionally, we also propose and discuss a couple of scenarios, specifically the programming of internet of things devices in a smart home environment, and check that the related configurations do not expose the user to potential external cyber-attacks. We will not provide any specific solution (with data and/or simulations), rather we want to discuss the potential collaboration between EUD and AI.