Open the Black Box: the Journey Toward Responsible Artificial Intelligence
Keywords: explainability, responsible AI, XAI
Abstract: Enabling AI systems to collaborate safely and efficiently with humans is one of the
ultimate goals in AI research. However, the majority of state-of-the-art intelligent
systems today operate as black-box models. This means that while we are aware
of their impressive capabilities, we remain unaware of the causal chains and
underlying logic driving their reasoning. This limitation hinders the advancement
of artificial intelligence and makes it challenging for humans to fully trust these
autonomous decision-making intelligent agents. Given these considerations, in
this essay, we offer a profound exploration of explainable artificial intelligence,
reviewing its fundamental paradigms, and contemplating potential pathways toward
responsible artificial intelligence in the future.
Submission Number: 218
Loading