Beyond explainability: justifiability and contestability of algorithmic decision systems

Beyond explainability: justifiability and contestability of algorithmic decision systems

In this paper, we point out that explainability is useful but not sufficient to ensure the legitimacy of algorithmic decision systems. We argue that the ke

When AI moves beyond human oversight: The cybersecurity risks of self-sustaining systems

Artificial intelligence is no longer just a tool executing predefined commands, it is increasingly capable of modifying itself, rewriting its own parameters, and evolving based on real-time feedback. This self-sustaining capability, sometimes referred to as autopoiesis, allows AI systems to adapt dynamically to their environments, making them more efficient but also far less predictable. For cybersecurity teams, this presents a fundamental chall…